Thiol-Ene Networks from Sequence-Defined Polyurethane Macromers

Article abstract of DOI:10.1021/jacs.0c00759

To date, scalability limitations have hindered the exploration and application of sequence-defined polymers in areas such as synthetic plastics, fibers, rubbers, coatings, and composites. Additionally, the impact of sequence on the properties of cross-lin

Full text of DOI:10.1021/jacs.0c00759

Subscriber access provided by University of Rochester | River Campus & Miner Libraries
Article
Thiol-Ene Networks from Sequence-Defined Polyurethane Macromers
Emily A. Hoff, Guilhem X. De Hoe, Christopher M. Mulvaney, Marc A. Hillmyer, and Christopher A. Alabi
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c00759 • Publication Date (Web): 23 Mar 2020
Downloaded from pubs.acs.org on March 23, 2020
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 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 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.
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 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Thiol-Ene Networks from Sequence-Defined Polyurethane Macromers
Emily A. Hoff1, Guilhem X. De Hoe2, Christopher M. Mulvaney1, Marc A. Hillmyer2, Christopher A.
Alabi1*
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
1Robert Frederick Smith School of Chemical & Biomolecular Engineering, 120 Olin Hall, Cornell University, Ithaca, NY
14835 USA
2Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455-0431, USA
ABSTRACT: To date, scalability limitations have hindered the exploration and application of sequence-defined polymers in
areas such as synthetic plastics, fibers, rubbers, coatings, and composites. Additionally, the impact of sequence on the
properties of cross-linked networks remains largely unknown. To address the need for synthetic methods to generate
sequence-defined materials in gram quantities, we have developed a strategy involving inexpensive and readily functional
vanillin-based monomers to assemble sequence-defined polyurethane oligomers via sequential reductive amination and
carbamation. Three oligomers were synthesized with monomer sequence precisely dictated by the placement of reactive side
chains during the reductive amination reaction. Avoiding excessive chromatographic purification and solid- or liquid-phase
supports enabled synthesis of sequence-defined oligomers on the gram-scale. Remarkably, sequence was shown to influence
network topology upon cross-linking as evidenced by sequence-dependent rubbery moduli values. This work provides one
of the first examples of a scalable synthetic route towards sequence-defined thermosets that exhibit sequence-dependent
properties.
group spacing, especially in rigid backbones, affect network
connectivity.
Introduction
Innovations in new polymeric material development
Following our previous studies with sequence-defined
oligothioetheramides (oligoTEAs) on the effect of backbone
have led to exciting and breakthrough technologies in fibers
and fabrics, high-value coatings, dental materials, personal
and military protection, and much more1,2. Polymeric
materials, such as polyurethanes, used in applications that
require high strength, solvent resistance, and high-
and pendant group sequence on biological properties7-10
,
we reasoned that sequence should also influence the
properties of networks used in materials science. We were
particularly inspired by recent reports from Johnson and
co-workers on unimolecular stereoisomeric block
copolymers (BCP) created via the Iterative Exponential
Growth (IEG) method at gram scale11. X-ray scattering
experiments showed that the stereochemical sequence
alone could be used to tune bulk morphology. Although IEG
is limited to repetitive or palindromic sequences, it allows
for gram-scale preparation and is the first platform to show
the influence of stereochemical sequence in unimolecular
polymers on bulk material morphology. Another example is
the pioneering work by Meyer and co-workers who
exploited the influence of sequence on the hydrolysis profile
of a synthetic, albeit polydisperse polymer, poly(lactic-co-
glycolic acid) (PLGA)12. In both examples, a process that
produced gram-scale amounts of material had to be
developed to examine the influence of sequence on self-
assembled copolymers and thermoplastics. Addressing the
question of how sequence affects the network connectivity
in thermosets, however, has not been done and requires
studying network formation with well-defined and easily
accessible functional macromolecules.
temperature
performance
traditionally
employ
thermosets1,3,4. Thermosets are classically made by cross-
linking either long polymer chains or multifunctional small
molecules. High cross-link densities and extensive
percolated networks are often responsible for high moduli
at service temperature and superior performance even
under harsh chemical environments1,5,6. In addition to
cross-link density, the distribution of cross-links, i.e.
network topology, is a strong determinant of material
property. Since most cross-linking reactions are kinetically
controlled (e.g., condensation and radical-based reactions),
formation of kinetically trapped states is dictated in part by
sterics and thus highly dependent on the geometry of the
multifunctional molecule or distribution of cross-linking
groups in the polymer chain. It follows that the relative
position or sequence of cross-linkable groups in a polymeric
chain, particularly in a rigid chain, should affect the kinetics
of cross-linking, the conformation of kinetically trapped
states, and thus the network structure and material
properties. However, due to the synthetic challenges
associated with creating precision polymers at scale, little is
known regarding how molecular geometry or functional
Creating synthetic polymeric chains with a dispersity of
one remains a major challenge in the field of polymer
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
chemistry. A variety of iterative addition techniques, which
involve addition of one monomer at a time to the end of a
growing polymer chain followed by purification, have been
employed to give
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
Figure 1. A) General synthesis scheme for scalable manufacturing of sequence-defined polyurethane macromers. (i) methanol,
Na2SO4, 2.5 h at RT, then NaBH4, 0 °C to RT in 1 h (ii) triethylamine, CH3CN, 50 °C, B) 1H NMR progression of SD-PUM synthesis, and
C) SD-PUM synthesis progression monitored via liquid chromatography.
short sequence-defined polymer compositions with a
dispersity of 113-23. Iterative addition on the solid phase via
insoluble beads facilitates rapid purification but can be
difficult to scale up due to surface area limitations.
Alternatively, several research groups, including our lab,
have proposed liquid-phase iterative synthetic methods to
bypass these limitations. These methods encompass the use
of “liquid polymeric/organic supports” or support-free
techniques that accomplish purification between each step
via chromatography techniques. Despite the advantages of
rapid solution phase kinetics, scale up of “liquid support”
systems is limited by significant purification losses and high
cost of fluorous tags or pre-polymer supports. Methods that
do not involve any supports are highly promising due to
their direct pathway for scale up. Few have explored this
path and notable recent examples include efforts by
Meier24,25, Johnson26,27, and Livingston28.
isomers, thus confirming the importance of sequence on the
properties of cross-linked materials. Access to this type of
investigation will contribute to fundamental design
principles for future applications in high-performance
materials.
Results and Discussion.
Synthesis
of
sequence-defined
polyurethane
macromers (SD-PUMs). We chose vanillin as the basis of
SD-PUM backbones because it is biorenewable, inexpensive,
possesses advantageous functional groups for orthogonal
synthesis, and will contribute to a relatively rigid polymeric
backbone. The benzylic aldehyde of vanillin reacts readily
with widely available primary amines to generate a useful
secondary amine handle. Furthermore, the phenol of
vanillin is easily converted to a reactive carbonate,
divanillin carbonate, on a large scale (~60 grams).
Carbamate backbone units were formed by reacting
divanillin carbonate with the secondary amine termini
generated via reductive amination.
Herein we describe a versatile strategy to synthesize
sequence-defined polyurethane macromers (SD-PUM)
derived from the biorenewable monomer, vanillin, via a
support-free strategy on the gram-scale. Key advantages of
this strategy include iterative assembly without a support,
inexpensive monomers, high-yielding reactions, and simple
liquid-liquid extraction for stepwise purification. Three
unique fluorescent SD-PUM isomers were created at the 2–
5 gram scale bearing reactive allyl and non-reactive methyl
pendant groups. Notably, the gram-scale synthesis of SD-
PUMs with precise sequence definition enabled the study of
structure-property relationships with respect to how
sequence affects the thermal and mechanical properties of
the final cross-linked networks. Our studies revealed that a
block SD-PUM sequence shows less favorable thermal and
mechanical properties relative to its other sequence
Preparation of each SD-PUM began by reacting 3,4-
dimethoxybenzaldehyde with the first desired primary
amine via a reductive amination reaction to generate a
secondary amine. After purification via a simple aqueous
wash, divanillin carbonate was then reacted with the
secondary amine to give a substituted carbamate with a
new aldehyde site which can be used for the next reductive
amination step (Fig. 1A). This carbamate formation
generates vanillin as a by-product, which is water soluble,
and the excess divanillin carbonate can be hydrolyzed to
vanillin using an ammonium hydroxide solution. An
alkaline-aqueous work-up is then sufficient to remove the
vanillin and ammonium hydroxide by-products. The
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
symmetric carbonate of vanillin was used for each
PUMs at scale with only a round-bottom flask, a separatory
funnel for purification, and a rotovap for solvent removal
for the majority of steps (Fig. 1A). In the last step of the
oligomer formation, divanillin carbonate was substituted
with a 4-nitrophenyl-activated
1
2
3
4
carbamation step (excluding the final capping carbamation)
because it yielded exclusively vanillin-derived carbamates.
An additional benefit of using this monomer is the potential
for recycling vanillin. Optimization of these synthesis and
purification steps enables the production of precision SD-
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
1
Figure 2. (A) SD-PUM structures and (B) H NMR of the isomeric 12-mers (C) retention time of SD-PUMs on a C4 reverse-phase
column (Supplementary Fig. 1 shows the spectra with an expanded baseline), (D) mass spectra of isomeric SD-PUMs (E) 2D EXSY
spectra at 200 ms mixing time for AAmmmA. (F) The off-diagonal peaks in the 200 ms EXSY spectra were used to calculate the rate
constants for conformer exchange. The Gibbs free energy of activation was calculated directly from the rate constant and the Eyring
equation.
carbonate of guaiacol, an analogue to vanillin without the
aldehyde, to provide a non-reactive end-group. The final
amination and carbamation steps were purified by column
chromatography.
at each step during preparation of mAmAmA are shown in
Fig. 1B. As shown in Fig. 1A, the 1-mer represents the
product of the first reductive amination step, i.e. 3,4-
dimethoxy-N-methylbenzylamine. After each carbamation
reaction with divanillin carbonate, the aldehyde peak at 10
ppm appears and after each reductive amination step, the
aldehyde peak disappears. Similarly, the resonance of the N-
methyl group in the N-methylbenzylamine adducts shifts
from 2.45 to 3 ppm (Fig. 1B, 1 vs. 2mer, 5 vs. 6mer and 9 vs.
10mer) and the methylene proton signals shift from 3.7 to
4.6 ppm upon carbamation (Fig. 1B). The synthesis process
was also monitored by LC-MS equipped with a diode-array
detector for simultaneous absorbance and mass analysis. A
consistent shift towards longer retention times with each
monomer addition (Fig. 1C) was observed. The absorbance
and molecular ion peak for each addition step corresponded
to the desired mass as shown in Supplementary Table 1. The
final yield of the three 12-mer isomers (Fig. 2A) ranged
from 7–10%. Even at these overall yields, sequence-defined
macromers were produced on a multigram scale with
inexpensive regents and minimal chromatographic
purification. As such, we present this synthetic strategy as a
scalable route to sequence-defined oligomers that enables
evaluation of sequence effects in applications (e.g. cross-
linked networks) where gram quantities are required.
The side chain sequence was varied by changing the order
of primary amines used in the reductive amination steps,
while the vanillin-based backbone was kept constant. So far,
methyl (non-reactive) and allyl (reactive) substituents have
been used as pendant groups to give three distinct,
monodisperse, and constitutionally isomeric sequence-
defined macromers in 2-5 gram quantities (Fig. 1A). Each
oligomer is named for its sequence of side chains, where m
represents a methyl substituent and A represents an allyl
substituent. The allyl functional groups were selected to
enable cross-linking of each oligomer via thiol-ene
chemistry. Three sequences were selected for investigation:
an alternating sequence (mAmAmA), a block sequence
(AAAmmm), and a comparative “disordered” sequence
(AAmmmA) (Fig. 1A). The modularity of the monomers and
reactions used in the preparation of SD-PUMs ensure that
this platform can be used to produce highly multifunctional
scaffolds in addition to the binary macromers in this report.
The progression of each SD-PUM synthesis was followed
by H NMR spectroscopy and liquid chromatography mass
spectrometry (LC-MS). As an example, the H NMR spectra
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
The 1H and 13C NMR spectra of all three SD-PUM
sequences shared similar spectral features with minor
differences in splitting patterns around the N-methyl
protons (Fig. 2B and Supplementary Fig. 2-4). The three
isomers also displayed a similar retention time (i.e. relative
hydrophobicity) when run on a reverse phase C4 column
(Fig. 2C), and all gave identical mass spectra (Fig. 1D) with
the desired molar mass ([M+H]+ = 1375.5 m/z, [M+NH4]+ =
1392.5 m/z , and [M+Na]+ = 1397.5 m/z).
Conformational exchange of SD-PUMs. While
monitoring the progress of oligomer synthesis via H NMR
1
1
2
3
4
5
6
7
8
spectroscopy, strong splitting patterns evolved for
backbone methylene and carbamate methyl peaks (Fig. 1B
and 2B) that suggested different conformational
populations of methylene and carbamate methyl groups
due to slow rotation around the C(O)-N bond. The two peaks
for the methyl group at ~3 ppm arise due to the differential
positioning of the protons on the methyl group
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 3. (A) Representative images of a SD-PUM that is then solvent cast and cross-linked via photoinitiated thiol-ene reaction with
[(2,2′-(ethylenedioxy)diethanethiol (EDET) or ethane-1,2-diyl bis(2-mercaptoacetate) (EBMA)] and the resulting free-standing film.
(B) Gel fraction obtained for each cross-linked oligomer at different UV-curing times. The 20- and 30-min time point measurements
were done in experimental duplicates while the 60-min time points were done in triplicate. The error bars represent the standard
deviation from the experimental replicates (three technical replicates per measurement). (C) Solution phase thiol-ene reaction
kinetics of SD-PUMs with 2-(2-ethoxyethoxy)ethane-1-thiol (EEET) monitored by the disappearance of allyl peaks at 5.23 and 5.86
ppm by 1H NMR spectroscopy. The lines drawn between the symbols are only meant to show the trends in the data. (D) H NMR
characterization of sol-extract from SD-PUM/EDET networks compared to pristine AAmmmA SD-PUM. The blue region highlights
the olefin peaks while the green region highlights the protons alpha to the urethane bond.
1
relative to the aromatic group, resulting in a more de-
shielded proton signal when the position is closer to the
aromatic group.
temperature NMR (VT-NMR) was also used to infer the
conformational dynamics of each SD-PUM. As the
temperature was increased from 25 to 100 °C, the
coalescence of the two methyl and methylene peaks at 4.6
and 3 ppm, respectively into one broader set of peaks for
each was indicative of hindered C-N rotation. The transition
occurred at 60 °C for all three sequences (Supplementary
Fig. 8-10). These results imply that the macromers exhibit
hindered rotation around the C(O)-N carbamate bond.
To probe the solution conformational dynamics of each
macromer, a two-dimensional homonuclear exchange
spectroscopy (1H-1H EXSY) experiment was conducted for
each oligomer at 0 ms and 200 ms mixing times. A
representative 2D NMR spectrum at 200 ms for AAmmmA
is shown in Fig. 2E. Complete 2D spectra for the other
macromers at 0 ms and 200 ms can be found in
Supplementary Fig. 5-7. The appearance of off-diagonal
EXSY peaks in the 200 ms spectra for each SD-PUM
confirmed the presence of slow conformational exchanges
between the different positions. The rate constants for
conformational exchange were calculated with the program
EXSYcalc by integrating the diagonal and EXSY peaks in
each spectrum. The Gibbs free energy of activation (ΔG‡) for
each SD-PUM was then determined via the Eyring equation
(Supplementary Eq. 1) using the forward rate constant and
was found to be ~16.3 kcal/mol across sequences (Fig. 1F).
This ΔG‡ is on par with values obtained for hindered
internal rotation in N,N-dialkyl amides29,30. Variable
Effect of sequence on network formation. Prior to
cross-linking, the thermal stability of each pure SD-PUM
was determined by thermogravimetric analysis (TGA)
(Supplementary Fig. 11). The thermal stability of each
oligomer, as assessed by mass loss, did not vary significantly
across sequences. Less than 5% mass loss was observed
below 280 °C, which is attributed in part to moisture and
solvent evaporation. To probe how sequence may affect
material properties, we exploited the allyl groups in the SD-
PUMs to create cross-linked networks. Networks were
prepared by solvent casting a mixture of each SD-PUM, a
dithiol cross-linker [(2,2′-(ethylenedioxy)diethanethiol
(EDET)
or
ethane-1,2-diyl
bis(2-mercaptoacetate)
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
(EBMA)], and
a
photoinitiator (2,2-dimethoxy-2-
reactions, each film was removed from the mold and
immersed in acetonitrile for 48 hours to remove
unincorporated monomers and photoinitiator from each
network. Once the soluble (sol) fractions were removed, the
films were dried in a vacuum oven for 16 hours at 100 °C.
1
2
3
4
5
6
7
8
phenylacetophenone (DMPA)) in acetonitrile into the
circular wells (100 mm in diameter x 1 mm deep) of a PTFE
mold. The dithiol cross-linkers were selected for their
commercial availability, miscibility with the SD-PUMs in
solution, and similar chain lengths to aid in network
comparisons. A representative image of a SD-PUM powder
(prior to solvent casting) cross-linked to give a freestanding
disc is shown in Fig. 3A. Cross-linking was initiated with UV
light (365 nm, 20 mW/cm2), and several curing times (20,
30, and 60 min) were investigated to provide kinetic
profiles of curing for each sequence. After the cross-linking
The strong UV absorption of the SD-PUMs (Fig. 2C) led us
to investigate their fluorescence emission upon excitation
with UV light. All three macromers exhibited intrinsic
fluorescence at 306 nm following excitation at 270 nm.
After generating calibration curves for each sequence,
fluorescence
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. Thermal and mechanical properties of sequence-controlled networks (60 min UV-cure). (A-D) DMTA of each cross-linked
network. The tan δ curves are shown for (A) the EDET cross-linked networks, and (B) EBMA cross-linked networks; dotted lines
correspond to the peak maxima for the AAAmmm and mAmAmA networks. (C) Plots of storage modulus (log scale) as a function of
temperature for EDET cross-linked networks; the inset shows a zoomed in view of the rubbery regime. (D) Plots of storage modulus
(log scale) as a function of temperature for EBMA cross-linked networks; the inset shows a zoomed in view of the rubbery regime.
measurements were used to determine the amount of
unincorporated SD-PUM, independent from the amount of
thiol-based cross-linker (reacted or unreacted) or initiator,
in the sol fraction at each curing time point; the
corresponding gel fractions were then calculated by
subtracting the sol fractions from the total amount of SD-
PUM (Fig. 3B). We confirmed in separate experiments that
the consumption of allyl groups via the thiol-ene reaction
did not significantly affect the fluorescence quantum yield.
Moreover, gel fraction data obtained by gravimetric
analysis were similar (albeit slightly higher due to the
contribution of the cross-linker mass) to those obtained by
fluorescence measurements (Supplementary Fig. 15). The
gel fraction data in Fig. 3B indicate that the AAmmmA and
AAAmmm sequences have slightly faster reaction kinetics
relative to the mAmAmA sequence. Reaction times below 20
min did not yield intact films and precluded the acquisition
of early stage gel fraction kinetics via the use of sol fraction
fluorescence measurements. To further explore the early
stage reaction kinetics, we performed control experiments
where the SD-PUMs were reacted with a monothiol, 2-(2-
ethoxyethoxy)ethane-1-thiol (EEET) under the same
conditions as the cross-linking reaction. The disappearance
1
of the allyl group was monitored by H NMR spectroscopy
(Supplementary Fig. 12-14) and the conversion was plotted
as a function of time (Fig. 3C). In this experiment, we
observed that the AAAmmm sequence had the fastest thiol-
ene reaction kinetics, followed by the AAmmmA, and the
mAmAmA sequence was again the slowest. We postulate
that oligomer conformation or a local concentration effect
is responsible for the faster reaction kinetics of the
AAAmmm sequence. To the best of our knowledge, thiol-
ene modification of sequence-defined systems such as ours
have not been demonstrated and computational studies are
underway to provide additional information to test the
1
aforementioned assertions. Due to peak overlap in the H
NMR spectra, it was not possible to reliably quantify thiol
conversion, but qualitatively the thiol peak at 1.6 ppm
ACS Paragon Plus Environment

Journal of the American Chemical Society
seemed to disappear at similar rate as the alkene peaks at
Page 6 of 9
is likely due to the variety of segmental motions in the
amorphous sample that results in broad endotherms. For
broad endotherms, small changes in heat capacity can be
difficult to detect and can obscure subtle differences
between samples. The tan delta (tan δ), or the ratio of E’ to
E”, depicts maxima during thermal transitions and is
therefore a common and often more sensitive way to define
the Tg of a material. To assess the Tg and mechanical
properties of our networks, larger films were prepared and
cut into rectangular tensile bars for dynamic mechanical
thermal analysis (DMTA) experiments (Supplementary Fig.
18). The tan δ plots from the DMTA experiments for each
network from are shown in Fig. 4A for the EDET networks
and Fig. 4B for the EBMA networks. The Tg obtained from
the tan δ maximum of the AAAmmm-EDET network (Tg ~
66 °C) was 9 °C and 4 °C lower than that of the mAmAmA-
and AAmmmA-EDET networks, respectively (Fig. 4A). A
similar trend was observed for the EBMA networks: the
AAAmmm-EBMA network had a Tg of 78 °C, which was 4 °C
lower than both mAmAmA- and AAmmmA-EBMA networks
(Fig. 4B).
1
2
3
4
5
6
7
8
5.2 and 6 ppm.
Although the SD-PUMs had minor differences in reaction
kinetics with EEET, all sequences reached a similar gel
fraction of 95-97 % after 60 min of irradiation (Fig. 3B).
With this near complete SD-PUM incorporation into the
network and relatively fast reaction kinetics (Fig. 3C), we
posited that all of the alkenes in the network were fully
reacted. To indirectly address this supposition, we probed
the sol fraction of the AAmmmA, mAmAmA, and AAAmmm
films after 60 min of curing with 1H NMR spectroscopy (Fig.
3D). The 1H NMR of the sol-fraction showed that in all cases,
virtually all allyl groups were reacted as evidenced by the
dearth of olefin peaks between 5.2 and 6 ppm (Fig. 3D,
highlighted blue region, less than 0.1% remaining at 5.2
ppm). We suspect the reacted oligomers in the sol fraction
are likely cyclic products or the result of having a small
excess of cross-linker in the reaction. Identification of these
individual structures using NMR spectroscopy and LC-MS
proved inconclusive. This result, along with the gel fraction
data in Fig. 3B, indirectly suggests that all three networks
have similar extents of cross-linking at the 60-min
irradiation time point. Following the gel-fraction data in Fig.
3B that indicated near complete network conversion after
60 min, additional networks using the EBMA cross-linker
were also cured for 60 min. Gel fraction analyses (by
fluorescence and gravimetric analysis) of these networks
also showed near complete SD-PUM incorporation
(Supplementary Fig. 16).
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
The storage modulus was measured as a function of
temperature for all six cross-linked films (Fig. 4C and 4D).
Below Tg, the three films made with the EDET cross-linker
had identical glassy moduli of 1.7 GPa. Similarly, the three
films made with the EBMA cross-linker had identical glassy
moduli of 2.3 GPa, approximately 26% stiffer than the EDET
networks. However, in the rubbery regime above Tg, the
network properties differed across each set of films. The
rubbery modulus for the AAAmmm-EDET network was
~40% and ~30% lower than that of the mAmAmA- and
AAmmmA-EDET networks, respectively (Fig. 4C, inset). We
surmise that these differences in mechanical properties
likely arise from factors influencing network topology, such
as loop formation and dangling chain ends that are further
discussed in the next section. A similar trend was revealed
for the EBMA networks (Fig. 4D, inset). Our DMTA findings
therefore indicated that both the thermal and mechanical
properties of the networks made with the block sequence
were distinct from those of the networks made from either
the alternating or “disordered” sequence. Additionally, a
slight increase in the rubbery modulus and Tg across all
sequences was observed for EBMA networks due to the less
flexible cross-linker when compared to the EDET networks.
Figure 5. Average storage modulus of each SD-PUM-EDET and
EBMA network at 150 °C from the DMTA experiments acquired
at a fixed frequency (6.28 rad/s, i.e. 1 Hz) with error bars
representing the standard deviation from the mean. The
“mixed sequence” is an equimolar ratio of the mAmAmA,
AAmmmA, and AAAmmm sequences. All measurements were
done in triplicate (i.e., measurement on three sample films)
except for the mAmAmA-EDET and AAAmmm-EDET networks
which were done in duplicates (for duplicates the bars
represent a range). **, p-value < 0.005, *, p-value < 0.05, ns p-
value > 0.05.
In addition to the temperature sweep DMTA data,
frequency and strain sweeps were also conducted for each
network at a fixed temperature of 150 °C (Supplementary
Fig. 19-22). The modulus values obtained from these tests
were in agreement with those obtained from the DMTA
temperature sweeps (Fig. 5 and Supplementary Fig. 23).
Regardless of the dithiol cross-linker used, a statistically
significant difference in modulus was observed between the
AAAmmm-based networks and the networks made from
the other two sequences (Fig. 5 and Supplementary Fig. 23).
To probe the role of sequence dispersity, we prepared a
“mixed sequence” containing an equimolar ratio (1:1:1) of
each SD-PUM and cured it for 60 minutes with the EDET
cross-linker. The rheological analyses revealed that the
modulus of the network made with the “mixed sequence”
was similar to the average modulus of all three pristine
networks (Fig. 5, green bar). The modulus of this “mixed
Effect of sequence on thermal and mechanical
properties of cross-linked networks. The thermal
properties of each EDET or EBMA cross-linked films were
evaluated by differential scanning calorimetry (DSC)
(Supplementary Fig. 17A and 17B). All sequences show
similar broad endotherms (Supplementary Fig. 17A). The
broadness of the endotherm, and thus Tg measured by DSC
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
sequence”-EDET network is still higher and statistically
different from that of the AAAmmm network. Thus, while
subtle, consistent property differences (mechanical and
thermal) between networks derived from AAAmmm and
those derived from mAmAmA, AAmmmA and the mixed
sequence highlight the role of sequence in dictating the
mechanical properties of precision thermoset materials.
conversion of the double bonds within the solid network. To
obtain direct information about the network conversion,
the EBMA networks containing acid labile ester linkages
were subjected to an acidic cleavage solution
(trifluoroacetic acid:acetonitrile:water, 1:1:1, v/v/v) for 12
h at 40 °C to achieve complete network degradation (Fig.
6A). Cleavage should result in SD-PUMs modified with one,
two, or three mercaptoacetic acid (MA) moieties (Fig. 6A
depicts the mAmAmA sequence modified with three MA
moieties) depending on the conversion of the thiol-ene
reaction.
1
2
3
4
5
6
7
8
Influence of sequence on network topology. While the
gel-fraction data and sol-fraction NMR experiments
suggested nearly complete network conversions were
achieved after 60 min (Fig. 3), neither directly examines the
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 6. (A) Schematic of model degradable network and a sample degradation fragment after treatment with (i) TFA:CH3CN:H2O
1:1:1 v/v/v mixture for 12 hours at 40 °C. (B) LC-MS characterization of degradation products after acid treatment. *Corresponds to
m/z of 1651 g/mol and **corresponds to m/z of 1557 g/mol. (C) Schematic of mAmAmA and AAAmmm networks labeled with the
different types of loops and defects.
The degraded networks were analyzed with LC-MS to
identify the extent of MA modification. The LC-MS analyses
showed that all sequences predominantly contained the
Fig. 24) with mAmAmA having relatively more species of
smaller masses than the other two sequences. These results
agree with the sol fraction analyses and support the
conclusion that the observed differences in thermal and
mechanical properties are related to the impact of sequence
on network connectivity, i.e. topology, and not the number
of cross-links. Recent reports by Johnson and coworkers on
loop formation in cross-linked networks31-34 suggest that
differences in inter- and intramolecular loop formation may
contribute to the differences we see in the AAAmmm
networks relative to the other two sequences. Since our gel
completely modified product (Fig. 6B, i.e.
3 MA
modifications, [M+H]+= 1651 m/z) suggesting near
complete conversion of the all double bonds within each
network. The mAmAmA sequence with the slowest kinetics
(Fig. 4B) showed a small amount of the doubly modified
product (Fig. 4D, ** [M+H]+= 1557 m/z). Additionally, the
sol fraction of the EBMA networks was degraded and also
showed predominantly modified product (Supplementary
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
Financial support for Emily Hoff, Christopher Mulvaney, and
fraction data and 2D NMR data suggest a similar degree of
cross-linking and conformational freedom, respectively, we
hypothesize that the observed differences in mechanical
behavior may be due to relative amounts of intramolecular
loops (Fig, 6C) and dangling chain ends. The presence of
elastically ineffective cross-links, such as loops, would
explain why we observe similar gel fraction behavior across
sequences, but differences in mechanical behavior. It is also
possible that the modulus of the AAAmmm networks is
lowered by the presence of dangling chain defects (i.e., the
mmm segment) throughout the network, as they would also
be elastically ineffective network defects35 (Fig. 6C).
Additionally, differences in overall oligomer conformation
could impact resultant network properties. Confirmation of
loop structures and conformational differences, however,
are beyond the scope of this initial work. Further studies are
required to ascertain the relative contributions of loop
structures and dangling chains to the network topology and
observed mechanical properties.
1
2
3
4
5
6
7
8
Christopher Alabi was provided by the Army Research Office
under award number W911NF-15-10179. Support for Guilhem
De Hoe and Marc Hillmyer was provided by the Center for
Sustainable Polymers under award number CHE-1413862.
Guilhem De Hoe also acknowledges the University of
Minnesota Doctoral Dissertation Fellowship. This work made
use of the Cornell University NMR Facility, which is supported,
in part, by the NSF under award number CHE-1531632. The
authors thank the Coates group for access to their DSC.
9
REFERENCES
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
1.
doi:10.1016/B978-0-85709-086-7.50013-4
2. Srinivasan, A. & Bandyopadhyay, S. Advances in Polymer
Materials and Technology. (CRC Press, 2016).
3. Pascault, J.-P., Sautereau, H., Verdu, J. & Williams, R. J. J.
Thermosetting Polymers. (CRC Press, 2002).
4. Thomas, S., Datta, J., Haponiuk, J. & Reghunadhan, A.
Guo, Q. Thermosets. (Woodhead Publishing, 2012).
Polyurethane Polymers: Blends and Interpenetrating Polymer
Networks. (Elsevier, 2017).
Conclusions.
5.
Auvergne, R., Caillol, S., David, G., Boutevin, B. & Pascault,
With the advent of synthetic sequence-defined polymers,
it has become necessary to address both the scalability of
sequence-defined materials and their application as tools to
better inform our understanding of materials design. In this
work, we report one of few examples of sequence-defined
oligomers that can be readily synthesized on a multigram-
scale. This was achieved through the combination of
efficient reactions, simple aqueous extractions, and
support-free synthesis. Cross-linking of alkene-functional
sidechains with different dithiols afforded networks with
sequence-dependent thermal and mechanical properties.
High conversion of the cross-linking reaction was
confirmed by gel-fraction and LC-MS analyses and the
sequence-dependent mechanical properties are the result
of topological differences related to the SD-PUM
microstructure. We propose that loop defects and dangling
chain-ends resulting from differences in cross-linkable
sidechain sequence are primarily responsible for the
observed property differences. This study is the first to
examine the role sequence-definition plays in network
topology and will pave the way for explorations of using
sequence to tune and control network properties.
J.-P. Biobased thermosetting epoxy: present and future. Chem. Rev.
114, 1082–1115 (2014).
6.
Publishing, 2014). doi:10.1016/B978-1-4557-3107-7.00023-3
7. Porel, M., Thornlow, D. N., Artim, C. M. & Alabi, C. A.
Handbook of Thermoset Plastics. (William Andrew
Sequence-Defined Backbone Modifications Regulate Antibacterial
Activity of OligoTEAs. Acs Chem Biol 12, 715–723 (2017).
8.
Hoff, E. A., Artim, C. M., Brown, J. S. & Alabi, C. A.
Sensitivity of Antibacterial Activity to Backbone Sequence in
Constitutionally Isomeric OligoTEAs. Macromol Biosci e1800241
(2018). doi:10.1002/mabi.201800241
9.
Brown, J. S. et al. Antibacterial isoamphipathic oligomers
highlight the importance of multimeric lipid aggregation for
antibacterial potency. Commun Biol 1, 220 (2018).
10.
Phan, N. N., Li, C. & Alabi, C. A. Intracellular Delivery via
Noncharged Sequence-Defined Cell-Penetrating Oligomers.
Bioconjugate. Chem. 29, 2628–2635 (2018).
11.
Golder, M. R. et al. Stereochemical Sequence Dictates
Unimolecular Diblock Copolymer Assembly. J. Am. Chem. Soc. 140,
1596–1599 (2018).
12.
Li, J., Stayshich, R. M. & Meyer, T. Y. Exploiting sequence
to control the hydrolysis behavior of biodegradable PLGA
copolymers. J. Am. Chem. Soc. 133, 6910–6913 (2011).
13.
Porel, M. & Alabi, C. A. Sequence-Defined Polymers via
Orthogonal Allyl Acrylamide Building Blocks. J. Am. Chem. Soc. 136,
13162–13165 (2014).
ASSOCIATED CONTENT
14.
Boeijen, A. & Liskamp, R. M. J. Solid‐Phase Synthesis of
Supporting Information
Oligourea Peptidomimetics. Eur. J. Org. Chem. 1999, 2127–2135
(1999).
15.
Synthesis and characterization procedures, characterization
data of all new compounds, and network data.
Edwardson, T. G. W., Carneiro, K. M. M., Serpell, C. J. &
Sleiman, H. F. An efficient and modular route to sequence-defined
polymers appended to DNA. Angew. Chem. Int. Ed. 53, 4567–4571
(2014).
The Supporting Information is available free of charge on the
ACS Publications website.
16.
controlled polymers. Science 341, 1238149 (2013).
17. Martens, S., Van den Begin, J., Madder, A., Prez, Du, F. E. &
Lutz, J.-F., Ouchi, M., Liu, D. R. & Sawamoto, M. Sequence-
AUTHOR INFORMATION
Espeel, P. Automated Synthesis of Monodisperse Oligomers,
Featuring Sequence Control and Tailored Functionalization. J. Am.
Chem. Soc. 138, 14182–14185 (2016).
Corresponding Author
* Correspondence to: caa238@cornell.edu
18.
Grate, J. W., Mo, K.-F. & Daily, M. D. Triazine-Based
Notes
Sequence-Defined Polymers with Side-Chain Diversity and
Backbone-Backbone Interaction Motifs. Angew. Chem. Int. Ed. 55,
3925–3930 (2016).
The authors declare no competing financial interest.
ACKNOWLEDGMENT
19.
Espeel, P. et al. Multifunctionalized Sequence-Defined
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
Oligomers from a Single Building Block. Angew. Chem. Int. Ed. 52,
13261–13264 (2013).
29.
Siddall, T. H., Stewart, W. E. & Knight, F. D. Nuclear
magnetic resonance studies of hindered internal rotation in higher
N,N-dialkylamides and -thionamides. J. Phys. Chem. 74, 3580–3583
(1970).
1
2
3
4
5
6
7
8
20.
Roy, R. K. et al. Design and synthesis of digitally encoded
polymers that can be decoded and erased. Nat Commun 6, 7237
(2015).
21.
30.
Kim, S. et al. Electronic Effect on the Molecular Motion of
Trinh, T. T., Oswald, L., Chan-Seng, D. & Lutz, J.-F.
Aromatic Amides: Combined Studies Using VT-NMR and Quantum
Calculations. Molecules 23, (2018).
31.
polymers from elastic solids to single-chain nanoparticles.
Chemical Science 10, 5332–5337 (2019).
32.
Proc. Natl. Acad. Sci. USA 109, 19119–19124 (2012).
33. Zhou, H. et al. Crossover experiments applied to network
Synthesis of molecularly encoded oligomers using
a
chemoselective ‘AB + CD’ iterative approach. Macromol Rapid
Commun 35, 141–145 (2014).
Wang, J. et al. Counting loops in sidechain-crosslinked
22.
Porel, M., Thornlow, D. N., Phan, N. N. & Alabi, C. A.
Sequence-defined bioactive macrocycles via an acid-catalysed
cascade reaction. Nat Chem 8, 590–596 (2016).
Zhou, H. et al. Counting primary loops in polymer gels.
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
23.
Kanasty, R. L. et al. Sequence‐Defined Oligomers from
formation reactions: improved strategies for counting elastically
inactive molecular defects in PEG gels and hyperbranched
polymers. J. Am. Chem. Soc. 136, 9464–9470 (2014).
Hydroxyproline Building Blocks for Parallel Synthesis
Applications. Angew. Chem. Int. Ed. 128, 9681–9685 (2016).
24.
Solleder, S. C., Zengel, D., Wetzel, K. S. & Meier, M. A. R. A
34.
Zhong, M., Wang, R., Kawamoto, K., Olsen, B. D. & Johnson,
Scalable and High-Yield Strategy for the Synthesis of Sequence-
Defined Macromolecules. Angew. Chem. Int. Ed. 55, 1204–1207
(2016).
J. A. Quantifying the impact of molecular defects on polymer
network elasticity. Science 353, 1264–1268 (2016).
35.
Hiemenz, P.C. & Lodge, T. P. Polymer Chemistry. (CRC
25.
conjugated oligomers. Sci Rep 8, 17483 (2018).
26. Barnes, J. C. et al. Iterative exponential growth of stereo-
and sequence-controlled polymers. Nat Chem 7, 810–815 (2015).
27. Leibfarth, F. A., Johnson, J. A. & Jamison, T. F. Scalable
Schneider, R. V. et al. Sequence-definition in stiff
Press, 2007).
synthesis of sequence-defined, unimolecular macromolecules by
Flow-IEG. Proc. Natl. Acad. Sci. USA 112, 10617–10622 (2015).
28.
Dong, R. et al. Sequence-defined multifunctional
polyethers via liquid-phase synthesis with molecular sieving. Nat
Chem (2018). doi:10.1038/s41557-018-0169-6
For table of contents only
ACS Paragon Plus Environment