Isotope Effects Reveal the Mechanism of Enamine Formation in l -Proline-Catalyzed α-Amination of Aldehydes

Article abstract of DOI:10.1021/jacs.5b10876

The mechanism of l-proline-catalyzed α-amination of 3-phenylpropionaldehyde was studied using a combination of experimental kinetic isotope effects (KIEs) and theoretical calculations. Observation of a significant carbonyl ¹³C KIE and a large p

Full text of DOI:10.1021/jacs.5b10876

Subscriber access provided by KUNGL TEKNISKA HOGSKOLAN
Communication
Isotope Effects Reveal Mechanism of Enamine Formation
in L-Proline Catalyzed #-Amination of Aldehydes
Melissa A Ashley, Jennifer S. Hirschi, Joseph A Izzo, and Mathew J. Vetticatt
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b10876 • Publication Date (Web): 15 Jan 2016
Downloaded from http://pubs.acs.org on January 16, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 6
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Isotope Effects Reveal Mechanism of Enamine Formation in
Proline Catalyzed αꢀAmination of Aldehydes
Lꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Melissa A. Ashley,† Jennifer S. Hirschi,† Joseph A. Izzo, and Mathew J. Vetticatt*
Department of Chemistry, Binghamton University, Binghamton, NY 13902.
diate (8) that serves as both an enolate equivalent and a
Brønsted acid activator of the electrophile, via proton transfer
from the carboxylic acid moiety, at the stereoꢀdetermining
transition state TS4 (Figure 1B). It is generally assumed that 8
forms via an intramolecular deprotonation mechanism from 7
via TS3. Oxazolidinone intermediates such as 11 and 13 are
considered offꢀcycle parasitic species within the HꢀL model.3
ABSTRACT: The mechanism of Lꢀproline catalyzed αꢀ
amination of 3ꢀphenylpropionaldehyde was studied using a
combination of experimental kinetic isotope effects (KIEs)
and theoretical calculations. Observation of a significant carꢀ
bonyl 13C KIE and a large primary (1°) α–deuterium KIE supꢀ
port rateꢀdetermining enamine formation. Theoretical predicꢀ
tions of KIEs exclude the widely accepted mechanism –
enamine formation via intramolecular deprotonation of an
iminium carboxylate intermediate (7). An E2ꢀelimination
mechanism catalyzed by a bifunctional base, that directly
forms an N-protonated enamine species (12•H+) from an oxaꢀ
zolidinone (11) intermediate, accounts for the experimental
KIEs. These findings provide the first experimental picture of
the transition state geometry of enamine formation and clarify
the role of oxazolidinones as nonꢀparasitic intermediates in
proline catalysis.
Seebach and Eschenmoser have proposed an alternate pathꢀ
way (SꢀE model), based on the observation of 11 and 13 by 1H
NMR.4 The key intermediate in the SꢀE pathway for enamine
catalysis is not 8 but synꢀenamine carboxylate 12. Formation
of 12 occurs via an E2ꢀelimination mechanism from 11. An
electrophileꢀinduced γꢀlactonization of 12 to oxazolidinone 13
via TS9 is the key stereoꢀdetermining event in this pathway
(Figure 1B). An NMR study by Gschwind and coꢀworkers5
supported a third mechanistic pathway – direct conversion of
11 to 8 (TSꢀGschwind) without the intermediacy of 7. The
Gschwind model combines key features of the HꢀL and SꢀE
mechanisms – apparent E2ꢀelimination from 11 (SꢀE proposal)
results in formation of HꢀL intermediate 8; and TS4 (HꢀL proꢀ
posal) is the stereoꢀdetermining event in the catalytic cycle.
The Lꢀproline catalyzed αꢀfunctionalization of aldehydes via
enamine catalysis has led to a number of powerful asymmetric
transformations.1 A single transition state model – the Houkꢀ
List model (HꢀL model) – provides a general rationale for the
observed enantioselectivity in these reactions (Figure 1).2 Cenꢀ
tral to this model is the anti-enamine carboxylic acid intermeꢀ
The three mechanisms discussed (Figure 1A) are different
with respect to the (a) mechanism of enamine formation, (b)
role of the oxazolidinone intermediate 11, or (c) nature of the
enantioselectivityꢀdetermining step (Figure 1C). While there is
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 6
little debate that TS4 (HꢀL model) is the stereoꢀdetermining
Theoretical studies. To aid in this quantitative interpretaꢀ
transition state,6 the exact mechanism of enamine formation or
the role of 11 in proline catalysis is not firmly established. We
report herein the results from a combined experimental and
tion of experimental KIEs, transition structures for each step in
Figure 1A were computed using the B3LYPꢀGD3 method12,13
employing a 6ꢀ31+G** basis set and a PCM solvent model14
for acetonitrile. This method adequately describes energetics
and predicts KIEs in other prolineꢀcatalyzed reactions.15 The
1
2
3
4
5
6
7
8
2
theoretical 13C and H kinetic isotope effect (KIE) study that
provides the first experimental insights into the transition state
geometry for enamine formation and clarifies the role of oxaꢀ
zolidinone intermediates in proline catalysis.
2
13C and H KIEs were computed from the scaled vibrational
frequencies of the respective transition structures using the
program ISOEFF98.16,17 A Wigner tunneling correction was
applied to all predicted KIEs.18
The mechanism of Lꢀproline catalyzed αꢀamination of aldeꢀ
hydes (Figure 1D)7 has been investigated using experimental8
and computational9 methods. Kinetic studies by Blackmond
have revealed that the reaction (a) is zero order in electrophile,
(b) exhibits asymmetric amplification, and (c) is autocatalytic.
Enamine formation has been implicated as the rateꢀ
determining step in the catalytic cycle. This reaction was
9
Table 1. Comparison of experimental and predicted KIEs for all tranꢀ
sition structures in Figure 1 not involved in enamine formation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
therefore chosen for determination of 13C and H KIEs as a
direct probe of the mechanism for enamine formation in catalꢀ
ysis by proline.
Experimental KIEs. Experimental 13C KIEs for 1a were
determined from analysis of starting material using NMR
methodology at natural abundance.10 Two separate reactions of
1a and 2a were taken to 84±2 % and 77±2 % conversion of
1a. Unreacted 1a was reꢀisolated from the reaction mixture
and the 13C isotopic composition compared to samples of unꢀ
reacted 1a, not subjected to reaction conditions.11 From the
changes in relative isotopic composition and the fractional
conversion, 13C KIEs were determined. Additionally, α–
deuterium KIEs (kHꢀ2/kDꢀ2) were measured, from two independꢀ
ent reactions of a 4:1 mixture of α–H2-1a:α–D2-1a, taken to
KIEs for steps not involved in enamine formation. A comꢀ
parison of experimental and predicted KIEs for all transition
structures in Figure 1A, excluding those involved in enamine
formation (TS3, TS8 and TSꢀGschwind), are shown in Table
1.19 A key observation is the poor match between all experiꢀ
mental and predicted KIEs for TS4 (HꢀL TS). This confirms
Blackmond’s finding7c that TS4 is not rateꢀdetermining. Preꢀ
dicted 13C KIEs at C1 for the remaining transition structures
(Table 1) are reasonably close to the experimental C1 KIE.
However, the corresponding predicted α–2H KIEs for these
transition structures are inverse – an unsurprising observation
considering the α–carbon is either uninvolved or completely
rehybridized in all these structures. Thus, the experimental 1°
α–deuterium KIE excludes all structures in Table 1 as the rateꢀ
determining step in catalysis.
2
48±2 % and 47±2 % conversion of α–H2-1a, using H NMR
analysis (of NaBD4 reduced reaction mixtures) to accurately
determine the enhancement of deuterium content in unreacted
starting material.11 Experimentally measured 13C and 2H KIEs,
from the four independent experiments are shown in Figure 2.
Figure 2. Experimental KIEs for Lꢀproline catalyzed reaction of 1a
2
with 2a. Two sets of 13C KIEs and two sets of H KIEs represent
independent experiments with six measurements per experiment.
Numbers in parentheses show the uncertainty in the last digit of each
measurement.
Qualitative interpretation of experimental KIEs. Obserꢀ
vation of a significant carbonyl (C1) 13C KIE and a large priꢀ
mary (1°) α–deuterium KIE (kHꢀ2/kDꢀ2) is indicative of a rateꢀ
determining step involving α–deprotonation concomitant with
bonding changes at C1. The small, yet nonꢀunity KIE on the αꢀ
carbon (C2) suggests that C2 is not completely rehybridized
during the proton transfer event. The experimental KIEs are
qualitatively consistent with a mechanism involving rateꢀ
determining E2ꢀelimination; however, a quantitative interpreꢀ
tation is deferred until all possible transition structures in the
various models (Figure 1A) are ruled out by a comparison of
predicted KIEs to experimental values.
Figure 3. Lowest energy transition structures for enamine formation
via deprotonation of 7 and comparison of experimental and predicted
KIEs19
Transition structures and KIEs for enamine formation. The
next step is to explore all possible transition structures for
enamine formation and compare the KIE predictions for each
structure to experiment. Intramolecular deprotonation of the
α–proton of 7 by the carboxylate moiety (TS3) is the proposed
2
ACS Paragon Plus Environment

Page 3 of 6
Journal of the American Chemical Society
mechanism of enamine formation in the HꢀL pathway. This bottom left corner of Figure 4 and shifts TS8Seebach towards an
1
2
3
4
5
6
7
8
mechanism, along with a waterꢀassisted conversion of 7 to 8
(TS3ꢀwat), has previously been studied computationally Figꢀ
ure 3).6 While the predicted normal α–2H KIEs for TS3 and
TS3ꢀwat are in crude agreement with experiment, the near
unity predicted 13C KIE on C1 is clearly inconsistent with
experiment (Figure 3). This result strongly rules against the
widely accepted notion that iminium carboxylate 7 is a direct
precursor to key enamine intermediate 8.
E2ꢀtype transition structure. Based on this reasoning, an alterꢀ
nate mechanism for direct formation of an enamine intermediꢀ
ate from 11 is proposed (Figure 5). In this new transition strucꢀ
ture TS8', a bifunctional acidꢀbase molecule protonates the
pyrrolidine nitrogen while simultaneously deprotonating the
α–proton of 11. The initial product from TS8' is Nꢀprotonated
syn or anti enamine carboxylate 12•H+ which can reꢀenter the
HꢀL pathway after a proton transfer to form 8 (Figure 5).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. More O’Ferrall–Jencks plot summarizing possible eliminaꢀ
tion pathways for the conversion of 11 to 12. The lower leftꢀhand
corner of the plot corresponds to zwitterionic intermediate 7, which
facilitates an E1ꢀelimination pathway.
Direct conversion of 11 to 12 (TS8Seebach) or 11 to 8 (TSꢀ
Gschwind) occurs via an elimination mechanism involving α–
deprotonation and CꢀO bond scission (Figure 4).20 Seebach has
proposed that an E2ꢀelimination pathway4 – represented by the
diagonal in the More O’Ferrall–Jencks plot for this transforꢀ
mation (Figure 4) – could be initiated by a number of bases
including another molecule of 4, 11, or even the product 3a
(autocatalysis). We tried modeling TS8Seebach (or TSꢀ
Gschwind) using these and other bases but all attempts to
locate an E2ꢀelimination transition structure resulted in geomꢀ
etries with the CꢀO bond completely cleaved as the base
deprotonated the α–proton. This corresponds to the second
step in a stepwise E1ꢀelimination mechanism proceeding via
the zwitterionic intermediate 7 – unsurprising, considering α–
deprotonation is more likely to occur from 7 than the less acidꢀ
ic 11. The magnitude of the predicted α–deuterium KIE for
TS8Seebach depends on the extent of deprotonation, which is a
function of the base employed for the particular calculation.
However, nearꢀcomplete CꢀO bond cleavage in all these strucꢀ
tures leads to closeꢀtoꢀunity values for the predicted C1 KIE –
an observation that is in clear disagreement with the ~2% exꢀ
perimental measurement.21
Figure 5. Transition structure (TS8'a) for an E2ꢀelimination mechaꢀ
nism consistent with experimental KIEs – most hydrogens have been
removed for clarity. Key bondꢀbreaking/making (black) and Hꢀ
bonding (red) distances (Å) shown along with predicted KIEs.
Several bifunctional bases22 were employed to model TS8'
and the best match of experimental and predicted KIEs was
obtained when soluble productꢀproline Hꢀbonded complex 15
was employed as the bifunctional base to effect the direct conꢀ
version of exo-11 to synꢀ12•H+. The key features of the resultꢀ
ing transition state geometry TS8'a, along with a comparison
of experimental and predicted KIEs, are shown in Figure 5.
Considering the complete mismatch between experimental and
theoretical KIEs for every transition structure modeled thus far
(TS1ꢀ10), the predicted values for TS8'a provide the best sim-
ultaneous match to all three key experimental measurements –
C1, C2, and α–deuterium KIEs. Finally, the calculated E+zpe
and free energy barriers for TS8'a are 14.1 and 28.0 kcal/mol,
respectively – values consistent with the facility of the reacꢀ
tion.23 These results strongly support E2-elimination from 11
as the most likely mechanism of enamine formation in proline
catalysis. We recognize that the α–deuterium KIEs for TS8'a
are predicted high in comparison to experimental values.19
Uncertainty regarding the exact identity of the base that cataꢀ
lyzes TS8' possibly accounts for this discrepancy. This inconꢀ
sistency could also arise from the known failure of calculaꢀ
tions based on conventional transition state theory (TST) in
accurately describing structures with concomitant heavy and
light atom motion.24 A variational transition state theory25
(VTST) treatment, for example, may give predictions that are
closer to experiment but the broad mechanistic picture that
emerges from such advanced calculations is expected to be
identical to the conclusions presented herein.
The conversion of 11 to 12 favors a stepwise E1ꢀelimination
mechanism over a concerted E2ꢀpathway due to stabilization
of the carbocation intermediate by the lone pair of electrons on
the pyrrolidine nitrogen in CꢀO bond cleavage. However, the
experimental KIEs point toward a concerted pathway. Disenꢀ
gaging the nitrogen lone pair from the reaction coordinate for
the elimination, by Hꢀbonding or protonation, destabilizes the
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 6
Our proposal, that 15 is likely the bifunctional base that catꢀ MJV acknowledges support from startup funding at Bingꢀ
1
2
3
4
5
6
7
8
alyzes the rateꢀdetermining step (TS8'), is consistent with (a)
Seebach’s proposal4 that base catalysis is the chemical origin
of autocatalysis observed in this reaction, i.e. product forꢀ
mation accelerates the reaction by increasing the concentration
of base that catalyzes the rateꢀdetermining step, and (b)
Blackmond’s observation that the autocatalytic nature of this
reaction is a result of ‘a catalytic cycle involving only soluble
proline complexes or soluble proline adducts.’8a After the origꢀ
inal submission of this manuscript, we were made aware of a
new NMR study by Gschwind and coꢀworkers probing the
mechanism of enamine formation in the proline catalyzed selfꢀ
aldol reaction of 3ꢀmethylbutanal in DMSO. This new study
rescinds their original proposal (Ref. 5, TSꢀGschwind) and
supports the HoukꢀList pathway (TS3) as the most likely
mechanism of enamine formation.26 This is in direct conflict
with our results (vide supra) and led us to further question the
conclusions presented in this manuscript.
We questioned whether our experimental KIEs resulted
from multiple steps in the catalytic cycle being partially rate
determining – for example, a weighted average of the predictꢀ
ed KIEs of TS2 and TS3 could potentially account for our
experimental KIEs. In order to probe this possibility experiꢀ
mentally, we determined the C1 KIEs using α–D2-1a as the
aldehyde. If TS3 was indeed partially rate determining, it is
expected that α–deuteriums would increase the barrier to TS3
and make it ‘more rateꢀdetermining’. This would result in a C1
KIE value closer to the predicted value for TS3 – 1.002 (Figꢀ
ure 3). We conducted duplicate 13C KIE experiments using α–
D2-1a and found the C1 KIE to be 1.024(4) and 1.021(4) –
virtually identical to our measurements using 1a.11 This result
confirms that our experimental KIEs originate from a single
rate-determining step and reaffirms that TS3 is not involved in
the mechanism of enamine formation in our system. The disꢀ
crepancy between our study and Ref. 26 is most likely atꢀ
tributable to the choice of electrophile (2a versus 3ꢀ
methylbutanal) and/or solvent (acetonitrile versus DMSO) for
the respective reactions.
hamton University from the SUNY Research Foundation and
the National Science Foundation through XSEDE resources
provided by the XSEDE Science Gateways program.
REFERENCES
9
(1) Science of Synthesis: Asymmetric Organocatalysis I; List, B.,
Maruoka, K., Eds.; Thieme: Stuttgart, Germany, 2012.
(2) Bahmanyar, S.; Houk, K. N.; Martin, H. J.; List, B. J. Am. Chem.
Soc. 2003, 125, 2475ꢀ2479.
(3) (a) List, B.; Hoang, Linh.; Martin, H. J. Proc. Nat. Acad. Sci.
2004, 101, 5839ꢀ5842. (b) Bock, D. A.; Lehmann, C. W.; List,
B. Proc. Nat. Acad. Sci. 2010, 107, 20636ꢀ20641.
(4) Seebach, D.; Beck, A. K.; Badine, D. M.; Limbach, M.; Eschenꢀ
moser, A.; Treasurywala, A. M.; Hobi, R.; Prikoszovich, W.;
Linder, B. Helv. Chim. Acta. 2007, 90, 425ꢀ471.
(5) Schmid, M. B.; Zeitler, K.; Gschwind, R. M. Angew. Chem. Int.
Ed. 2010, 49, 4997ꢀ5003.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(6) Sharma, A. K.; Sunoj, R. B. Angew. Chem. Int. Ed. 2010, 49,
6373ꢀ6377.
(7) List, B. J. Am. Chem. Soc. 2002, 124, 5656ꢀ5657.
(8) (a) Iwamura, H.; Wells, D. H.; Mathew, S. P.; Klussmann, M.;
Armstrong, A.; Blackmond, D. G. J. Am. Chem. Soc. 2004, 126,
16312ꢀ16313. (b) Iwamura, H.; Mathew, S. P.; Blackmond, D.
G. J. Am. Chem. Soc. 2004, 126, 11770ꢀ11771. (c) Mathew, S.
P.; Klussmann, M.; Iwamura, H.; Wells, D. H.; Armstrong, A.;
Blackmond, D. G. Chem. Comm. 2006, 4291ꢀ4293.
(9) Sharma, A. K.; Sunoj, R. B. Chem. Comm. 2011, 47, 5759–5761.
(10) Singleton, D. A.; Thomas, A. A. J. Am. Chem. Soc. 1995, 117,
9357ꢀ9358.
(11) See Supporting Information for detailed description of the experꢀ
iments involved in determination of KIEs.
(12) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(13) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
(14) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999ꢀ3093.
(15) Zhu, H.; Clemente, F. R.; Houk, K. N.; Meyer, M. P. J. Am.
Chem. Soc. 2009, 131, 1632ꢀ1633.
(16) Anisimov, V.; Paneth, P. J. Math. Chem. 1999, 26, 75.
(17) Frequencies were scaled by 0.9614; Scott, A. P.; Radom, L. J.
Phys. Chem. 1996, 100, 16502.
(18) Bell, R. P. The tunnel effect in chemistry; Chapman & Hall: Lonꢀ
don, 1980.
(19) These transition structures and predicted KIEs were calculated
using multiple DFT methods. See Supporting Information for
geometries and full details of these calculations.
In conclusion, this work resolves the mechanism of enamine
formation in the proline catalyzed αꢀamination of aldehydes.
Our data supports a mechanism involving direct conversion of
oxazolidinone 11 to Nꢀprotonated enamine 12•H+ via an E2ꢀ
elimination initiated by a bifunctional base. Rapid proton
transfer from 12•H+ presumably forms 8 followed by reꢀentry
into the HoukꢀList pathway. These results confirm the role of
oxazolidinone 11 as a key non-parasitic intermediate in the
HoukꢀList catalytic cycle while invoking base catalysis as the
possible origin of autocatalysis observed in this reaction.
(20) To model TSꢀGschwind, a Brønsted acid (AcOH) is added to
TS8 to stabilize the incipient carboxylate moiety.
(21) See Supporting Information for detailed discussion
(22) Since the exact identity of the bifunctional base that catalyzed
TS8' is unknown, transition structures corresponding to TS8'
were located using a wide variety of bifunctional bases. See
Supporting Information for full details of these calculations.
(23) See Supporting Information for calculated energy barriers for
TS8'a using several DFT methods. For an important discussion
on interpretation of calculated free energy barriers see: Plata, R.
E.; Singleton, D. A. J. Am. Chem. Soc. 2015, 137, 3811ꢀ3826.
(24) Vetticatt, M. J. PhD. Dissertation, Texas A&M University, 2009.
(25) Truhlar, D. G.; Garrett, B. C. Acc. Chem. Res. 1980, 13, 440–
448.
ASSOCIATED CONTENT
Supporting Information. Complete experimental and compuꢀ
tational details, and NMR data are included in the Supporting
Information. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author vetticatt@binghamton.edu
†
These authors contributed equally to this work.
(26) Haindl, M. H.; Hioe, J.; Gschwind, R. M. J. Am. Chem. Soc.
2015, 137, 12835ꢀ12842.
ACKNOWLEDGMENT
4
ACS Paragon Plus Environment

Page 5 of 6
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
30x16mm (600 x 600 DPI)
ACS Paragon Plus Environment