Single-Molecule Investigation of Initiation Dynamics of an Organometallic Catalyst

Article abstract of DOI:10.1021/jacs.6b00357

The action of molecular catalysts comprises multiple microscopic kinetic steps whose nature is of central importance in determining catalyst activity and selectivity. Single-molecule microscopy enables the direct examination of these steps, including eluc

Full text of DOI:10.1021/jacs.6b00357

Subscriber access provided by UNIV OF PITTSBURGH
Article
Single-Molecule Investigation of Initiation
Dynamics of an Organometallic Catalyst
James D Ng, Sunil P Upadhyay, Angela N Marquard, Katherine M Lupo, Daniel
A Hinton, Nicolas A Padilla, Desiree M. Bates, and Randall H. Goldsmith
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b00357 • Publication Date (Web): 04 Mar 2016
Downloaded from http://pubs.acs.org on March 6, 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 11
Journal of the American Chemical Society
1
2
3
4
5
6
Single-Molecule Investigation of Initiation Dynamics of an Organo-
metallic Catalyst
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
James D. Ng, Sunil P. Upadhyay, Angela N. Marquard, Katherine M. Lupo, Daniel A. Hinton, Ni-
colas A. Padilla, Desiree M. Bates, Randall H. Goldsmith*
Department of Chemistry, The University of Wisconsin-Madison, 1101 University Avenue, Madison, Wisconsin,
53706, United States
ABSTRACT: The action of molecular catalysts comprises multiple microscopic kinetic steps whose nature is of central
importance in determining catalyst activity and selectivity. Single-molecule microscopy enables the direct examination of
these steps, including elucidation of molecule-to-molecule variability. Such molecular diversity is particularly important
in the behavior of molecular catalysts supported at surfaces. We present the first combined investigation of the initiation
dynamics of an operational palladium cross-coupling catalyst at the bulk and single-molecule levels, including under
turn-over conditions. Base-initiated kinetics reveal highly heterogeneous behavior indicative of diverse catalyst popula-
tion. Unexpectedly, this distribution becomes more heterogeneous at increasing base concentration. We model this be-
havior with a two-step saturation mechanism and identify specific microscopic steps where chemical variability must exist
in order to yield observed behavior. Critically, we reveal how structural diversity at a surface translates into heterogeneity
in catalyst behavior, while demonstrating how single-molecule experiments can contribute to understanding of molecular
catalysts.
INTRODUCTION
neity, and describe a potential mechanism that suggests
possible chemical origins of that diversity. By providing a
unique, new perspective on catalyst behavior, these re-
sults help us understand the subtle differences between
the action of a catalyst in solution as compared to a sur-
face, and constitute important steps in the evolution of
SM techniques to tackle chemical dynamics.
Single molecule (SM) fluorescence microscopy is a power-
ful method for the observation of unsynchronized chemi-
cal dynamics and nanoscale behavior. While biological
1,2
structure and dynamics are rich targets for SM measure-
3-5
ments, non-biological applications of SM techniques
6-9
have
interactions,
photo-induced electron transfer,
explored
diffusion,
chromatographic
10,11
12
adsorption and desorption behavior,
An ideal molecular catalyst for SM studies would have
several qualities facilitating SM fluorescence microscopy.
The ideal catalyst must be reliably tethered to a surface,
and immobilized in the microscope sample plane. The
test system must be fluorescently labeled in a way that
allows the fluorescence signal to convey information on
the catalyst’s chemical transitions. Finally, a good catalyst
for study would be stable for long periods of time in am-
bient conditions. For the above reasons, pyridine en-
hanced precatalyst, preparation, stabilization, and initia-
tion (PEPPSI) palladium catalysts are an attractive target
for SM studies. PEPPSI catalysts are employed in a variety
of coupling chemistries including Suzuki-Miyaura reac-
13,14
15
and photocatalysis.
A particularly rich abiotic application of SM microscopy
has been investigation of chemical reaction kinetics at
nanoparticle catalysts, including gold and copper na-
noparticles, platinum-silica core-shell particles, and in-
organic nanocrystals. A variety of nanoparticle reaction
16
17
18
19
types have been explored, including acid and base cataly-
20,21
22-24
sis
catalysts, like that of enzymes,
and redox catalysis.
Reactivity of nanoparticle
has been observed to be
More recently, the action of
25,26
1
6,18,19,27
highly heterogeneous.
28-30
small-molecule chemical agents
and organometallic
3
1,32
complexes
have also begun to be explored at the SM
3
3
34-37
tions,
37,38
39
level. Study of molecular catalysts by SM fluorescence
microscopy promises to improve our understanding of
chemical reactions by providing insight into catalyst ki-
netics and heterogeneity within catalyst populations. This
work details a combined SM fluorescence and bulk NMR
investigation into the initiation dynamics of an opera-
tional surface-supported molecular palladium catalyst. In
the process we are able to correlate SM fluorescence dy-
namics with specific processes in the cascade of chemical
steps comprising catalyst initiation, elucidate the diverse
dynamics that result from the intrinsic surface heteroge-
as well as amination
and sulfonylation reac-
tions, due to a combination of superior air and moisture
3
5
40
stability and high catalyst activity. PEPPSI catalysts
provide a convenient synthetic handle for reliable surface
4
1
support through the N-heterocyclic carbene and for ad-
dition of fluorescent functionality via the labile pyridine
4
2
ligand. Importantly, as loss of the pyridine ligand must
3
6
precede formation of the active catalyst, and as the lig-
and itself has been implicated in the remarkable stability
3
4
of PEPPSI catalysts, a fluorescent label at this ligand
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
offers the opportunity to explore these important pro-
cesses at the SM level.
mode. In TIRF, laser excitation is restricted to close prox-
imity to the surface, making our fluorescence measure-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
7
ments highly surface selective. In this experimental con-
figuration (Figure 1c), BODIPY fluorophores appended to
uninitiated catalyst will be fluorescent, while ligands
freed during initiation will quickly diffuse out of the exci-
tation volume and no longer emit, making fluorescence
intensity a targeted probe of catalyst initiation. By con-
trolling the surface density of catalyst molecules so that
we can observe individual catalysts on the surface, we can
gather kinetic information on initiation resolved at the
level of single catalyst molecules (Figure 1c, bottom).
In order to probe the initiation mechanism of the
PEPPSI palladium catalyst through SM fluorescence mi-
croscopy, we have synthesized a surface supported
PEPPSI palladium catalyst functionalized with a boron-
dipyrromethene (BODIPY) fluorophore appended to the
pyridine ligand, 1, Figure 1a. The fluorescent moiety pro-
vides a proxy for initiation by reporting on the presence of
pyridine ligated to the palladium. BODIPY fluorophores
are an attractive label for molecular catalysts due to their
synthetic modularity, stability, high fluorescent quantum
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
43
yield, and lack of reactive nucleophilic heteroatoms,
Understanding the mechanism and kinetics of initia-
tion of surface-supported molecular catalysts is an out-
standing problem in catalysis, where supported catalysts
have long been discussed as potential recyclable catalysts
which allows the fluorophore to act as a benign spectator
molecule near organometallic complexes (Supporting
44
Information). We have specifically chosen a spirofluo-
Figure 1. Single-molecule investigation of surface-supported molecular catalysts. A surface-supported, BODIPY la-
belled catalyst (a), or surface-supported BODIPY model compound (b), is deposited on a glass sample chamber, and
viewed on a microscope. Multiple catalyst molecules are observed simultaneously (c, top left). The catalysts fluoresce
(
c, top) until one initiates, at which time the fluorescent ligand leaves the excitation volume and signal from that cata-
lyst is lost (c, top middle). As multiple catalysts initiate, the fluorescence intensity traces (c, bottom, 400uM KOt-Bu)
from each will show a transition, providing initiation kinetics resolved at the level of individual molecules. The sur-
face anchoring chemistry in (a) is shown differently for the two molecules, representing diverse bonding configura-
rene functionalized BODIPY with an expanded π-
due to their ease of recovery. Though surface-catalyst
4
5
48,49
system as our fluorescent reporter because it enables use
of red excitation, substantially reducing background pho-
interactions can provide new avenues for reactivity,
surface-supported catalysts typically exhibit less favorable
reaction kinetics or selectivity than their solution-phase
*
tons from solvent Raman scattering and nꢀπ transitions
4
6
50
of impurities in non-aqueous solvents, and the stiff spi-
ro- groups greatly improve its quantum yield relative to
analogs. Quantifying initiation kinetics of surface-
supported molecular catalysts is particularly challenging
because the surface heterogeneity and biphasic condi-
tions result in substantial obstacles to spectroscopy, in-
cluding NMR. The rich chemical environment at a solid-
liquid interface can create diverse speciation of supported
catalysts that has been linked to diverse chemical activi-
4
5
other red fluorophores. As will be shown, this red
BODIPY demonstrates excellent imaging properties for
SM microscopy in non-aqueous protic media like 2-
propanol (IPA).
In order to observe initiation of 1 under Suzuki-
Miyaura coupling conditions, we conduct wide-field fluo-
rescence microscopy employing an objective-type pseudo-
total internal reflectance fluorescence (TIRF) excitation
5
1
ty, while sluggish initiation kinetics may play a role in
the reduced activity of surface-supported catalysts. Con-
sequently, understanding the influence of the surface on
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
resulted in the formation of free pyridine ligand and deu-
catalyst behavior is critical to improvement of supported
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
molecular catalysts.
terated acetone, as well as biphenyl product (Figure 2,
Supporting Information). The proposed initiation mecha-
nism is shown in Figure 2c. The IPA solvent is deproto-
RESULTS AND DISCUSSION
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Initiation of 5 in deuterated IPA, reaction, and mechanism during Suzuki cross-coupling. Dissociation of the
pyridine ligand from the catalyst was studied under turnover conditions (a), and upon exposure to individual reagents.
Ligand dissociation results in a deshielding of the pyridinic doublets as monitored by NMR (b). NMR spectra of solutions
of homogeneous catalyst (5) and individual reagents (3, 4), as well as the full reaction mixture show that KOtBu is both
necessary and sufficient for dissociation of pyridine from 5, consistent with an initiation mechanism driven by attack by
deprotonated solvent and ending in β-hydride elimination (c). See Supporting Information for full NMR of reaction mix-
NMR study of initiation of a homogeneous catalyst
nated by base, the isopropoxide ion displaces a chloride
ligand on the catalyst, thus lowering the activation barrier
to pyridine ligand dissociation by nearly 3 kcal/mol (Sup-
porting Information). This dissociation opens up a vacant
coordination site for β-hydride elimination of isopropox-
ide to form acetone. Reductive elimination of HCl forms
The initiation of 1 under Suzuki-Miyaura coupling con-
ditions is likely a multi-step process
5
2,53
0
ending with a Pd
species capable of entering the well-known cross-coupling
catalytic cycle. In order to isolate the role of each reagent
on catalyst initiation under typical coupling conditions
0
3
4
the Pd active catalyst. Isopropoxide coordination fol-
(Figure 2a), we examined solutions containing a homo-
geneous, unlabeled catalyst and various combinations of
reagents by NMR (Figure 2b). Addition of the reaction
substrates, 3 and 4, to the catalyst showed no shift in the
NMR spectrum, indicating ligand dissociation did not
occur. Addition of potassium tert-butoxide (KOtBu)
caused the disappearance of the peaks corresponding to
the pyridine bound to the catalyst, and the appearance of
peaks corresponding to the free ligand. Pyridine dissocia-
tion occurred rapidly and was complete by 15 seconds
after addition of KOtBu. Under these same conditions,
deuterated acetone formation was observed after several
hours, indicative of β-hydride elimination of isopropoxide
(Supporting Information). Addition of all reagents also
lowed by β-hydride elimination to form acetone has been
suggested as a key step in the activation of other Pd-NHC
5
4
complexes in IPA.
Single-molecule initiation kinetics of surface-supported
catalyst
As summarized in Figure 2c and exhibited in our NMR
measurements, release of the pyridine ligand into solu-
tion is a critical step along the initiation cascade, and can
be used to monitor the initiation kinetics of 1 at the SM
level. A sparse layer of 1 embedded in a dense layer of
triethoxy(octyl)silane (TOS) was prepared on a silica sur-
4
2
face in a specially designed all-glass reaction chamber
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
containing IPA, to form a sparse distribution of catalyst in to the surface support with the same anchoring function-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a high surface coverage of the alkyl silane. TOS slows the
base-induced destruction of the silica support by creating
a non-polar barrier to base attack. A large volume of IPA
with KOtBu of known molarity with or without reaction
substrates is quickly added, and time-lapse fluorescence
images are recorded over time (example in Supplemen-
ality as 1, under similar conditions to 1. The median sur-
vival times of 2 were found to be 1-2.5 orders of magni-
tude greater than those of 1 (Figure 3c), indicating a sig-
nificant separation of timescales of catalyst initiation
(fast) and combined destruction of the surface support
and photobleaching (slow). The mixing time of solution
5
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Extraction of SM kinetics. In an automated process, time/fluorescence trajectories from individual cata-
lysts (a, black) are isolated from video of a sparse catalyst-coated surface, and the frame where the fluorescence is
lost (blue line) is found for each. (b) The surviving fraction S(t), aggregated from initiation of many (>1000) mole-
cules, shows conspicuously stretched exponential behavior at high [KOtBu], visible as an elbow in the decay. Medi-
an lifetimes (which reflect the broad distribution of component rates, see Supporting Information) of 1 and 2 were
-1
well separated across the concentration range studied (c). Median lifetimes are roughly linear vs. [KOtBu] (c, in-
st
set), indicative of 1 order behavior. Because the concentration of catalyst is small compared to concentration of
base, [KOtBu] is expected to remain constant over the course of the experiment. Traces in (a) taken from a single
-1
video of 400 µM base addition. The 10 M lifetime of the model compound has a large uncertainty due to our
observation window being less than half of the median lifetime. The error bar for this point was omitted for clarity
of the logarithmic plot.
tary Information). We use a home-built drift compensa-
tion system to maintain focus during and after the addi-
tion of reagent (further information in Methods and Sup-
plementary Information). The red spiro-functionalized
BODIPY is seen to be a stable SM emitter under these
conditions (Figure 3a, Supporting Information). Individu-
al molecules of 1 were identified and survival times were
determined using an automated algorithm (Figure 3a).
The remaining fraction of surviving molecules, S, was
plotted as a function of time (Figure 3b) for a series of
increasing concentrations of KOtBu. Higher concentra-
tions of base result in conspicuously non-exponential de-
in the sample chamber, blinking of the BODIPY fluoro-
phore, and presence of fluorescent impurities were all
considered as possible confounding processes in our ex-
periment, and determined to be minor contributors (Sup-
porting Information). Consequently, catalyst initiation
can be viewed as the dominant mechanism of fluores-
cence loss, and fluorescence loss can be used as a proxy
for catalyst initiation.
The highly non-monoexponential kinetics of catalyst
5
6,57
initiation were investigated by Weibull analysis,
a
technique widely used in reliability engineering, to model
S with the stretched exponential form shown in equation
(1).
cay kinetics. Median survival times (t ) are linear with
1/2
the inverse of base concentration (Figure 3c, inset), indic-
ative of a first order decay process.
ꢍ
ꢀꢁꢂꢃ ꢄ ꢅꢆꢇꢈ
ꢉꢊꢊꢋꢌ
(1)
(2)
While loss of the fluorescently labeled ligand due
to catalyst initiation will result in loss of the SM fluores-
cence signal, other loss mechanisms must also be consid-
ered. In order to ascertain the contributions of photo-
bleaching, as well as lysis of the surface anchoring func-
tionality, which could also result in loss of fluorescence,
we measured 2, in which the BODIPY is covalently bound
ꢒ
lnꢁꢎlnꢁꢀꢁꢂꢃꢃꢃ ꢄ ꢏꢐln ꢂ ꢎ ꢏ ln ꢑ
ꢓ
ꢈꢉꢊꢊ
In Weibull analysis, S is linearized on a Weibull plot as
equation (2), and the shape parameter β and scale param-
eter (effective rate constant) keff of the Weibull distribu-
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
tion are extracted from the slope and intercept of the lin-
ear fit, respectively. Monoexponential kinetics are re-
vealed as lines with a slope of one (Figure 4a, dashed grey
line), while stretched exponential behavior is indicated by
β<1, with lower β values easily seen as lines approaching
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 4. Analysis of stretched exponential behavior with
the Weibull distribution. Survival fractions after base
addition to catalyst in IPA (a, X) are well-fit by lines with
slope less than one (- - -) on a Weibull plot (a), indicative
of stretched exponential behavior. A line on the Weibull
plot with slope one (grey - - -) would indicate a monoex-
ponential decay. Deviation from linearity at early times
in the low concentration data accounts for <10% of cata-
lyst molecules (right axis). Examination of the shape
parameter (slope), β, for the catalyst with base (b) and
catalyst under turn-over conditions (c) show similar
trends of decreasing β, suggesting similar drivers of het-
erogeneity. The trend in β indicates a broadening distri-
bution of rates (d, calculated from β in b). At low β (red),
both extremely fast and extremely slow rates are more
common than in distributions with high β (black). At β=1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
grey - - -), the distribution of exponential rates becomes
a delta function (monoexponential) at keff (inset d).
horizontal and indicative of increasingly broad distribu-
tions of rates. The distribution of survival times of 1 after
addition of KOtBu in IPA (Figure 4a) is well approximated
by the linearized fit shape. β decreases with increasing
base concentration both when only base is added (Figure
4b) and under turnover conditions (Figure 4c). Decay
dynamics of 2 also show a decreasing shape parameter
(Supporting Information), albeit at substantially reduced
rate constants. Though catalysts show the same qualita-
tive trend (Figure 4b,c) with and without the inclusion of
the arylhalide 3 and arylboronic acid 4, we will focus our
analysis on the base addition without additional reagents
due to a lower concentration of fluorescent impurities.
Additional discussion of behavior in the presence of 3 and
4 is presented as Supporting Information.
Stretched exponential kinetics can be viewed as deriv-
ing from a continuous distribution, H(k), of monoexpo-
58,59
nential processes.
ꢍ
ꢘ
ꢅꢆꢇꢈꢉꢊꢊꢋꢌ ꢄ ꢔ ꢕꢁꢖꢃꢅꢆꢋꢈꢗꢖ
(3)
ꢙ
Smaller values of β indicate a greater dispersion of rate
constants. The inverse relationship between base concen-
tration and β indicates increasing spread in the relative
rate constant distribution and greater heterogeneity of
catalyst behavior with addition of base. This changing
distribution, H(k), can be directly visualized (Figure 4d),
with H(k) approaching a singular value as β approaches
unity.
Physical interpretation of stretched decays
Surface-supported molecular catalysts inhabit envi-
ronments significantly more complex than their homoge-
neous analogs due to the variety of surface microenvi-
ronments. On silica support, this variety begins with the
5
1
distribution of silanol sites across the silica surface, lead-
ing to a wide range of possible attachment configurations
for the surface-supported catalyst. In addition, while the
catalyst is extremely dilute during the deposition, the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
high concentration of TOS in the deposition solution
could allow silane groups to coalesce and condense ho-
the dense foliage of alkane surrounding the catalyst, satu-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ration behavior will be observed (functionally equivalent
60
16
mogeneously, leading to deposition at the surface of a
variety of large catalyst-silane clusters, largely non-polar
layers that may pose a significant obstacle to the polar
isopropoxide. The range of attachment cluster sizes and
geometries, as well as the range of attachment chemis-
tries of those clusters, leads to a highly polymorphous
catalyst microenvironment. Moreover, a variety of con-
formations adopted by the catalysts’ flexible alkyl tether
can lead to a variety of orientations relative to the surface.
Because these microenvironments influence the iso-
propoxide base’s approach to and interaction with the
catalyst metal center, the surface heterogeneity will be
reflected in the catalyst behavior, including initiation.
Critically, though the existence of structural heterogenei-
ty at the surface has been verified by other experimental
to a Langmuir-Hinshelwood mechanism or Michaelis-
26
Menten kinetics, which have both been explored at the
SM level),
(4)
ꢈ ꢟꢠꢡ ꢟꢣꢡ
ꢞ
ꢢ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢚꢛꢜꢝ ꢄ ꢤ
(5)
ꢧꢤ
ꢥꢦ
ꢤ
ꢞꢨꢟꢣꢡ
ꢦ
where k
diffuses into and out of sites (A) in the alkane pocket
surrounding the catalyst with total number of sites [A]
is the rate constant of initiation of the precatalyst com-
and k-1 are the rates at which isopropoxide (B)
1
,
T
k
2
5
1
approaches, our investigation constitutes the first direct
observation of the effect of the molecular catalyst struc-
tural heterogeneity on the catalyst dynamics. That effect
is encapsulated in the increasing spread of rates in Figure
4d.
plex once the isopropoxide enters the pocket, and Robs is
the observed reaction rate. As will be shown below, equa-
tion (5) provides a means of generating our observed ki-
netic distributions, with heterogeneity arising naturally as
some species are limited by the first step (k ) while others
1
are limited by the second step (k ). Under these condi-
2
That a diversity of microenvironments would yield a
corresponding diversity of initiation dynamics is antici-
pated. The unexpected element of Figure 4 is that the
apparent diversity increases as a function of base concen-
tration. If the heterogeneous dynamics stem from the
heterogeneity in microenvironments, then a single under-
lying structural distribution, largely independent of base
concentration, must be capable of yielding increasingly
heterogeneous dynamics under different chemical condi-
tions. Said another way, the structural diversity may be
nearly constant in all samples, but the dynamics appear to
be increasingly heterogeneous at high base concentra-
tions, despite likely deriving from the same underlying
family of structures.
tions, the order of Robs with respect to base will be first
order at low base concentrations, and become zeroth or-
der at high concentration as the average occupancy of the
pocket increases. However, with singular values for all of
the component rate constants, a Weibull plot will show a
series of parallel lines with slope, β =1 (monoexponential
decay), with curves coalescing as saturation is approached
(Supporting Information), contrary to our observations.
In order for the slope of the Weibull plot to decrease
with increasing base concentration, as observed, there
must be a distribution of maximum rates among catalysts.
In such a system, the distribution of observed rates will
broaden as an increasing fraction of the catalysts ap-
proach saturation, leading to a change in the slope of the
associated Weibull plot. Critically, a distribution of values
of only certain parameters in equation (5) are capable of
providing the observed rate dependence, thus reporting
on the microscopic origins of the macroscopic diversity.
Conceptually, we can imagine the changing distributions
deriving from equation (5) as a transitioning of catalysts
operating in the high vs. low concentration regime. At
high concentrations of base, the model approaches equa-
tion (6),
Saturation kinetics of the catalyst can give rise to the
increasingly wide distribution of rates we observe. In a
saturation model, the presence of a diversity of maximum
rates will yield highly monodisperse rates below the satu-
ration point, but a broadening distribution of rates as
individual catalysts near saturation. In this mechanism, a
single underlying distribution of saturation points can
yield a variety of narrow or broad distributions of behav-
iors. The geometry of the surface is extremely conducive
to the creation of saturation kinetics. Molecular dynamics
simulations of alkane functionalized silica surfaces have
shown a highly disordered network of surface-supported
alkanes, with pockets of low alkane density as well as
ꢚꢛꢜꢝ ꢄ ꢖ ꢟꢪꢡ
(6)
ꢩ
ꢫ
while at low concentration, the saturation model ap-
proaches equation (7), for small k-1
61
dense patches. Pockets in the immediate vicinity of cata-
.
lyst molecules constitute positions proximal to the cata-
lyst available for occupation by polar isopropoxide mole-
cules, with pockets of different size leading to differences
in the number of total available positions, while the posi-
tion and shape of the pockets can modulate base activity.
Because isopropoxide molecules (B) must first occupy
these entry sites (A) before reacting with the catalyst and
releasing pyridine (ie, a two-step mechanism), and be-
cause there are a limited number of sites available due to
ꢚꢛꢜꢝ ꢄ ꢖ ꢟꢪꢡ ꢟꢬꢡ
(7)
ꢒ
ꢫ
A distribution of values of k would lead to behavior
1
opposite to that observed, with a broad distribution of
rates at low base concentrations (eq. 7) that narrows at
high concentrations (eq. 6). In contrast, a distribution of
k2 will lead to a broadening distribution of rates for large
[B]. However, because the distribution does not exhibit
monoexponential behavior at any concentration over the
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
range studied (Figure 5), a distribution of either [A] or k1
is also required. Simulations generated from distributions
tributions in the parameters generate the observed distri-
T
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
butions in our kinetics, in this same regime, the rate will
st
in k and k did not generate the magnitude of change in β
still be 1 order in [B], preserving the observed bulk be-
1
2
seen in our data over a similar concentration range, sug-
gesting distributions of [A] and k are responsible for the
havior with regard to base, (Figure 5). At substantially
higher concentrations, the process shifts to zero order in
base, consistent with equation (6) (Figure 5b, inset).
T
2
observed catalyst kinetics (Figure 5). Critically, while dis-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Simulation of survival times with distributions in k and [A] . Simulated data (a, solid) qualitatively match
2
T
experimental data (X), capturing the inverse relationship between slope (β) and concentration. Due to the double-log y-
axis, points below -2 account for <10% of the population. In the simulation, a marginally greater range of concentrations
is needed to observe the change in lifetime in our data (purple line). First order dependence on [KOtBu] is preserved
over the critical range in the simulation (b), albeit with a different slope. Deviations are most conspicuous at low
[
KOtBu]. Closer agreement between the observed and simulated data for both (a) and (b) could be obtained by also in-
cluding distributions of values for k and k . Extension of the simulated bulk kinetics to substantially higher base con-
1
-1
centrations shows the expected saturation behavior (b, inset). Simulations were generated based on equation (5) with k
1
-1
-1
-8 -1
=
20 s M and k = 1x10 s . A Weibull distribution (τ = 3.5, β = 0.9) was randomly sampled to generate [A] , as well as k
-1 T
2
(
τ = 2, β = 0.9). Other types of distributions, as well as distributions in other variables, led to poorer fits (Supporting In-
We interpret k as relating to variations in catalyst rate
precatalyst changes from site to site (both reflected in k ).
2
2
constant due to changes in the local microenvironment,
This model of hindered attack can also explain the chang-
and [A] as a measure of the maximum attainable local
ing shape parameter in the model 2 (Supporting Infor-
mation), where base-assisted destruction of the silyl ether
attachment group will depend on accessibility of the sup-
port surface to isopropoxide in the same way the catalyst
kinetics depend on accessibility of the metal. Because the
majority of observed heterogeneity can be generated with
T
concentration of isopropoxide at the palladium center of
1. This maximum local concentration is controlled by the
crowding of the local catalyst environment by TOS. Be-
cause the contour length of TOS is longer than the con-
tour length of the tether on 1, the alkyl chains can be ex-
pected to rise above the metal center of the catalyst. The
environment around each catalyst will be heterogeneous
and greasy, with local density of alkyl chains creating a
unique solvent compartment around each catalyst. This
range of local geometries can create differences in perme-
a distribution in [A] alone (Supporting Information), we
T
view barriers to isopropoxide attack due to a limited
number of available entry sites for isopropoxide, as the
main driver of catalyst heterogeneity.
CONCLUSIONS
ability of the alkane layer (k ), differences in the maxi-
1
We have described the first combined SM and bulk in-
vestigation of organometallic catalyst initiation. The initi-
ation kinetics were observed to be highly heterogeneous,
grew more heterogeneous with increasing base concen-
mum achievable local concentration of base ([A] ), and
T
also differences in accessibility of the metal center to base
due to steric effects and differences in activated complex
stability as the steric crowding of the bulky activated
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
tration, and this increasing heterogeneity was shown to
Supporting Information. Additional experimental detail;
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
absorption and emission spectra of BODIPY labeled mole-
cules 1 and 2; discussion of possible confounders including
solution mixing, camera frame rate, single molecule blinking,
solvent impurities; catalyst behavior in presence of coupling
reagents; robustness of fit parameters; method of data calcu-
lation; NMR of numbered compounds; Example time-lapse
video. This material is available free of charge via the Inter-
net at http://pubs.acs.org.
be consistent with a saturation model that includes the
need to access a limited number of entry sites. More
broadly, population diversity in catalysts has important
implications to a wide range of reactions, where inactive
subpopulations could significantly slow bulk reaction
kinetics. For example, our results indicate that this highly
heterogeneous distribution of observed rates at high base
concentrations includes a significant population of slow-
initiating species, even as the majority initiates quickly.
To better characterize the way in which the attachment
chemistry affects the catalyst heterogeneity, catalysts
linked by more rigid or longer tethers, could be compared
to this study. Further study of catalyst initiation could
provide insight into causes of catalyst inefficiency and
failure by allowing us to create a more detailed model of
supported catalyst behavior that includes this diversity.
Our approach is applicable to a wide variety of surface-
supported catalysts, including electrocatalysts, and will be
particularly powerful when paired with a means of moni-
AUTHOR INFORMATION
Corresponding Author
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*rhg@chem.wisc.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by the National Science Foundation
under Grant No. (CHE-1254936). We gratefully acknowledge
partial support of this research by NSF through the Universi-
ty of Wisconsin Materials Research Science and Engineering
Center (DMR-1121288), and would like to thank Margaret
Robinson for invaluable help with surface attachment strate-
gy.
1
6,25
toring individual turnover events,
where the presence
or absence of specific ligands can be correlated with reac-
tivity. Use of a FRET labeled catalyst would enable inde-
pendent observation of both ligand and catalyst, poten-
tially allowing observation of ligand return dynamics.
Insight into truly homogeneous catalyst behavior will
require careful control of surface-catalyst interactions to
reduce surface influence, a process we are currently ex-
ploring. In total, SM techniques hold the potential to be
powerful new addition to the toolkit for mechanistic in-
vestigation of organometallic species.
REFERENCES
(
1)
Janssen, K. P.; De Cremer, G.; Neely, R. K.;
Kubarev, A. V.; Van Loon, J.; Martens, J. A.; De Vos, D. E.; Roeffaers,
M. B.; Hofkens, J. Chem. Soc. Rev. 2014, 43, 990.
(2)
3)
Moerner, W. E. Angew. Chem. Int. Ed. 2015, 54,
8
067.
(
Joo, C.; Balci, H.; Ishitsuka, Y.; Buranachai, C.;
Methods
Ha, T. Annu. Rev. Biochem. 2008, 77, 51.
4) Xie, X. S.; Choi, P. J.; Li, G. W.; Lee, N. K.; Lia, G.
Ann. Rev. Biophys. 2008, 37, 417.
(
Synthesis
Full synthetic details are given in the Supporting In-
(5)
Deniz, A. A.; Mukhopadhyay, S.; Lemke, E. A. J.
62
formation. 4-ethynylpyridine was obtained 90% yield
R. Soc. Lond. Interface 2008, 5, 15.
4
5
and the brominated red BODIPY fluorophore was ob-
tained in 38% yield by following literature procedures.
Sonogashira coupling between the alkyne and aryl halide
(6)
Zurner, A.; Kirstein, J.; Doblinger, M.; Brauchle,
C.; Bein, T. Nature 2007, 450, 705.
(
7)
Liao, Y.; Yang, S. K.; Koh, K.; Matzger, A. J.;
Biteen, J. S. Nano. Lett. 2012, 12, 3080.
(
35% yield), followed by hydrogenation (58% yield) yield-
(8)
2006, 110, 18939.
(9)
Werley, C. A.; Moerner, W. E. J. Phys. Chem. B
4
2
ed the PEPPSI ligand. Following an earlier procedure,
compound 1 was prepared from the PEPPSI ligand and a
functionalized Pd dimer by refluxing in DCM at elevated
temperature (44% yield).
Higgins, D. A.; Park, S. C.; Tran-Ba, K. H.; Ito, T.
Ann. Rev. Anal. Chem. 2015, 8, 193.
10) Kisley, L.; Chen, J.; Mansur, A. P.; Shuang, B.;
(
Kourentzi, K.; Poongavanam, M. V.; Chen, W. H.; Dhamane, S.;
Willson, R. C.; Landes, C. F. Proc. Natl. Acad. Sci. U.S.A. 2014, 111,
2075.
Fluorescence Microscopy
Full details on the experimental setup are given in the Sup-
porting Information. We performed SM fluorescence micros-
copy on an inverted microscope with 633 nm excitation pro-
vided in a pseudo-TIRF geometry. An electronic shutter
chopped excitation light for time-lapse imaging. The camera
recorded a 100 ms exposure every 3 s. An active drift com-
pensation routine was used to dampen focal shifts upon rea-
gent addition and maintain focus for prolonged periods by
mapping focal drift onto the spatial displacement of the re-
flected TIRF beam as it leaves the back aperture of the mi-
croscope as observed on a camera.
(11)
Cooper, J. T.; Peterson, E. M.; Harris, J. M. Anal.
Chem. 2013, 85, 9363.
(
12)
Kastantin, M.; Walder, R.; Schwartz, D. K.
Langmuir 2012, 28, 12443.
(13)
Rao, V. G.; Dhital, B.; Lu, H. P. Chem. Comm.
2015, 51, 16821.
(
(
14)
15)
Jin, S.; Lian, T. Nano Lett. 2009, 9, 2448.
Tachikawa, T.; Yamashita, S.; Majima, T. J. Am.
Chem. Soc. 2011, 133, 7197.
(16)
Mater. 2008, 7, 992.
(17)
Xu, W.; Kong, J. S.; Yeh, Y. T.; Chen, P. Nat.
Decan, M. R.; Impellizzeri, S.; Marin, M. L.;
Scaiano, J. C. Nat. Commun. 2014, 5, 4612.
18) Han, R.; Ha, J. W.; Xiao, C.; Pei, Y.; Qi, Z.; Dong,
B.; Bormann, N. L.; Huang, W.; Fang, N. Angew. Chem. Int. Ed. 2014,
3, 12865.
(
ASSOCIATED CONTENT
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
(
19)
Schryver, F. C.; Jacobs, P. A.; De Vos, D. E.; Hofkens, J. Nature 2006,
39, 572.
Roeffaers, M. B.; Sels, B. F.; Uji, I. H.; De
(40)
Larrosa, I.; Somoza, C.; Banquy, A.; Goldup, S.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
M. Org. Lett. 2011, 31, 146.
4
(41)
Phan, N. T. S.; Van Der Sluys, M.; Jones, C. W.
(
20)
Roeffaers, M. B. J.; De Cremer, G.; Libeert, J.;
Adv. Synth. Catal. 2006, 348, 609.
Ameloot, R.; Dedecker, P.; Bons, A.-J.; Bückins, M.; Martens, J. A.;
Sels, B. F.; De Vos, D. E.; Hofkens, J. Angew. Chem. Int. Ed. 2009, 48,
9285.
(21)
De Vos, D. E.; Hofkens, J.; Roeffaers, M. B. J. ACS Nano 2014, 8,
2650.
(42)
Upadhyay, S. P.; Lupo, K. M.; Marquard, A. N.;
Ng, J. D.; Bates, D. M.; Goldsmith, R. H. J. Phys. Chem. C 2015, 119,
19703.
Liu, K.-L.; Kubarev, A. V.; Van Loon, J.; Uji-i, H.;
(43)
4891−4932.
(44)
Loudet, A.; Burgess, K. Chem. Rev. 2007,
1
Canham, S.; Bass, J.; Navarro, O.; Lim, S.-G.;
(
22)
Ristanović, Z.; Kerssens, M. M.; Kubarev, A. V.;
Das, N.; Blum, S. A. Organometallics 2008, 27, 2172.
(45) Kowada, T.; Yamaguchi, S.; Ohe, K. Org. Lett.
2010, 12, 296.
Hendriks, F. C.; Dedecker, P.; Hofkens, J.; Roeffaers, M. B. J.;
Weckhuysen, B. M. Angew. Chem. Int. Ed. 2015, 54, 1836.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(23)
Martínez, V. M.; Cremer, G. D.; Roeffaers, M. B.
(46)
2003, 74, 3597.
(47)
Moerner, W. E.; Fromm, D. P. Rev. Sci. Instrum.
J.; Sliwa, M.; Baruah, M.; De Vos, D. E.; Hofkens, J.; Sels, B. F. J. Am.
Chem. Soc. 2008, 130, 13192.
Martin-Fernandez, M. L.; Tynan, C. J.; Webb, S.
(
24)
G. Q.; Nissen, E. J.; Thomas, E. M.; Zou, N. M.; Chen, P. Nature 2016,
30, 77.
Sambur, J. B.; Chen, T. Y.; Choudhary, E.; Chen,
E. J. Microsc. 2013, 252, 16.
(48)
Notestein, J. M.; Katz, A. Chem. Eur. J. 2006, 12,
5
3954.
(25)
English, B. P.; Min, W.; van Oijen, A. M.; Lee, K.
(49)
Canlas, C. P.; Lu, J.; Ray, N. A.; Grosso-
T.; Luo, G.; Sun, H.; Cherayil, B. J.; Kou, S. C.; Xie, X. S. Nat. Chem.
Biol. 2006, 2, 87.
Giordano, N. A.; Lee, S.; Elam, J. W.; Winans, R. E.; Van Duyne, R. P.;
Stair, P. C.; Notestein, J. M. Nat. Chem. 2012, 4, 1030.
(
26)
Lu, H. P.; Xun, L.; Xie, X. S. Science 1998, 282,
Xu, W.; Kong, J. S.; Chen, P. J. Phys. Chem. C
Rybina, A.; Thaler, B.; Kramer, R.; Herten, D. P.
(50)
Madhavan, N.; Jones, C. W.; Weck, M. Acc.
1
877.
Chem. Res. 2008, 41, 1153.
(
27)
(51)
Coperet, C.; Chabanas, M.; Saint-Arroman, R.
2009, 113, 2393.
28)
P.; Basset, J.-M. Angew. Chem. Int. Ed. 2003, 42, 156.
(
(52)
Steinhoff, B. A.; Guzei, I. A.; Stahl, S. S. J. Am.
Phys. Chem. Chem. Phys. 2014, 16, 19550.
29) Rybina, A.; Lang, C.; Wirtz, M.; Grussmayer, K.;
Chem. Soc. 2004, 126, 11268.
(
(53)
Schultz, M. J.; Adler, R. S.; Zierkiewicz, W.;
Kurz, A.; Maier, F.; Schmitt, A.; Trapp, O.; Jung, G.; Herten, D. P.
Angew. Chem. Int. Ed. 2013, 52, 6322.
Primalov, T.; Sigman, M. S. J. Am. Chem. Soc. 2005, 127, 8499.
(54)
Singh, R.; Viciu, M. S.; Kramareva, N.; Navarro,
(30)
Ameloot, R.; Roeffaers, M.; Baruah, M.; De
O.; Nolan, S. P. Org. Lett. 2005, 7, 1829.
Cremer, G.; Sels, B.; De Vos, D.; Hofkens, J. Photochem. Photobiol.
Sci. 2009, 8, 453.
(55)
(56)
Schunk, T. C. J. Chromatogr. A 1993, 656, 289.
Abernethy, R. B.; Breneman, J. E.; Medlin, C. H.;
(
31)
Esfandiari, N. M.; Wang, Y.; Bass, J. Y.; Blum, S.
Reinman, G. L. Weibull Analysis Handbook, United Technologies
Corporation, 1983.
A. Inorg. Chem. 2011, 50, 9201.
(
32)
Esfandiari, N. M.; Blum, S. A. J. Am. Chem. Soc.
(57)
Flomenbom, O.; Velonia, K.; Loos, D.; Masuo,
2011, 133, 18145.
S.; Cotlet, M.; Engelborghs, Y.; Hofkens, J.; Rowan, A. E.; Nolte, R. J.;
Van der Auweraer, M.; de Schryver, F. C.; Klafter, J. Proc. Natl. Acad.
Sci. U.S.A. 2005, 102, 2368.
(33)
(
Cordes, T.; Blum, S. A. Nat. Chem. 2013, 5, 993.
Nasielski, J.; Hadei, N.; Achonduh, G.; Kantchev,
34)
E. A.; O'Brien, C. J.; Lough, A.; Organ, M. G. Chem. Eur. J. 2010, 16,
0844.
(58)
1980, 73, 3348.
(59)
Lindsey, C. P.; Patterson, G. D. J. Chem. Phys.
1
(
35)
Dash, C.; Shaikh, M. M.; Ghosh, P. Eur. J. Inorg.
Berberan-Santos, M. N.; Bodunov, E. N.; Valeur,
Chem. 2009, 2009, 1608.
B. Chem. Phys. 2005, 315, 171.
(36)
O'Brien, C. J.; Kantchev, E. A.; Valente, C.;
(60)
(61)
Witucki, G. L. J. Coating. Tech. Res. 1993, 65, 57.
Lippa, K. A.; Sander, L. C. J. Chromatogr. A
Hadei, N.; Chass, G. A.; Lough, A.; Hopkinson, A. C.; Organ, M. G.
Chem. Eur. J. 2006, 12, 4743.
2006, 1128, 79.
(62)
(
37)
Kirschning, A.; Mennecke, K. Synthesis 2008,
Moreno-Garcia, P.; Gulcur, M.; Manrique, D. Z.;
2
008, 3267.
Pope, T.; Hong, W.; Kaliginedi, V.; Huang, C.; Batsanov, A. S.; Bryce,
M. R.; Lambert, C.; Wandlowski, T. J. Am. Chem. Soc. 2013, 135, 12228.
(38)
Organ, M. G.; Abdel-Hadi, M.; Avola, S.;
Dubovyk, I.; Hadei, N.; Kantchev, E. A.; O'Brien, C. J.; Sayah, M.;
Valente, C. Chem. Eur. J. 2008, 14, 2443.
(
39)
Sayah, M.; Organ, M. G. Chem. Eur. J. 2011, 17,
1
1719.
ACS Paragon Plus Environment

Page 11 of 11
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
11
ACS Paragon Plus Environment