nucleation-elongation dynamics of two-dimensional …
TRANSCRIPT
doi.org/10.26434/chemrxiv.8309339.v1
Nucleation-Elongation Dynamics of Two-Dimensional Covalent OrganicFrameworksHaoyuan Li, Austin Evans, Ioannina Castano, Michael Strauss, William Dichtel, Jean-Luc Bredas
Submitted date: 21/06/2019 • Posted date: 24/06/2019Licence: CC BY-NC-ND 4.0Citation information: Li, Haoyuan; Evans, Austin; Castano, Ioannina; Strauss, Michael; Dichtel, William;Bredas, Jean-Luc (2019): Nucleation-Elongation Dynamics of Two-Dimensional Covalent OrganicFrameworks. ChemRxiv. Preprint.
Homogeneous two-dimensional (2D) polymerization is a poorly understood process in which topologicallyplanar monomers react to form planar macromolecules, often termed 2D covalent organic frameworks(COFs). While these COFs have traditionally been limited to weakly crystalline aggregated powders, theywere recently grown as micron-sized single crystals by temporally resolving the growth and nucleationprocesses. Here, we present a quantitative analysis of the nucleation and growth rates of 2D COFs via kineticMonte Carlo (KMC) simulations, which show that nucleation and growth have second-order and first-orderdependences on monomer concentration, respectively. The computational results were confirmedexperimentally by systematic measurements of COF nucleation and growth rates performed via in situ X-rayscattering, which validated the respective monomer concentration dependences of the nucleation andelongation processes. A major consequence is that there exists a threshold monomer concentration belowwhich growth dominates over nucleation. Our computational and experimental findings rationalize recentempirical observations that, in the formation of 2D COF single crystals, growth dominates over nucleationwhen monomers are added slowly, so as to limit their steady-state concentration. This mechanisticunderstanding of the nucleation and growth processes will inform the rational control of polymerization in twodimensions and ultimately enable access to high-quality samples of designed two-dimensional polymers.
File list (2)
download fileview on ChemRxivCOFNucleationGrowth_062119_ChemRxiv.pdf (0.90 MiB)
download fileview on ChemRxivCOFNucleationGrowth-SI-062119_ChemRxiv.pdf (3.59 MiB)
1
Nucleation-Elongation Dynamics of Two-
Dimensional Covalent Organic Frameworks
Haoyuan Lia§, Austin M. Evansb§, Ioannina Castanob, Michael J. Straussb, William R. Dichtelb*
and Jean-Luc Bredasa*
aSchool of Chemistry and Biochemistry, Center for Organic Photonics and Electronics (COPE),
Georgia Institute of Technology, Atlanta, Georgia 30332-0400
bDepartment of Chemistry, Northwestern University, Evanston, IL 60208, USA.
§These authors contributed equally to this work.
Corresponding Authors
2
ABSTRACT
Homogeneous two-dimensional (2D) polymerization is a poorly understood process in which
topologically planar monomers react to form planar macromolecules, often termed 2D covalent
organic frameworks (COFs). While these COFs have traditionally been limited to weakly
crystalline aggregated powders, they were recently grown as micron-sized single crystals by
temporally resolving the growth and nucleation processes. Here, we present a quantitative analysis
of the nucleation and growth rates of 2D COFs via kinetic Monte Carlo (KMC) simulations, which
show that nucleation and growth have second-order and first-order dependences on monomer
concentration, respectively. The computational results were confirmed experimentally by
systematic measurements of COF nucleation and growth rates performed via in situ X-ray
scattering, which validated the respective monomer concentration dependences of the nucleation
and elongation processes. A major consequence is that there exists a threshold monomer
concentration below which growth dominates over nucleation. Our computational and
experimental findings rationalize recent empirical observations that, in the formation of 2D COF
single crystals, growth dominates over nucleation when monomers are added slowly, so as to limit
their steady-state concentration. This mechanistic understanding of the nucleation and growth
processes will inform the rational control of polymerization in two dimensions and ultimately
enable access to high-quality samples of designed two-dimensional polymers.
3
TOC GRAPHICS
KEYWORDS: Covalent organic frameworks (COFs); two-dimensional (2D) materials; kinetic
Monte Carlo simulations; two-dimensional polymer networks; in situ X-ray diffraction
4
Introduction
Two-dimensional (2D) polymerization is an emerging frontier in polymer science, which promises
the development of rationally designed 2D materials.1 2D covalent organic frameworks (COFs)
are a class of 2D polymers that are permanently porous, atomically regular, synthetically modular,
and structurally versatile.2-14 These properties have inspired explorations of COFs for applications
such as catalysis, molecular separations, energy storage, and optoelectronic devices.15-34 However,
their small crystalline domain sizes (< 100 nm) and empirical polymerization processes have
impeded the study of their intrinsic properties. It has been suggested that the insufficient materials
quality of the 2D COFs obtained to-date is likely due to uncontrolled nucleation-elongation
processes in the reaction mixture.35-38 However, characterizing the small, polydisperse, constantly
evolving particles present in the early stages of COF polymerization remains a demanding
experimental task. From a theoretical perspective, using an analytical approach to describe the
nucleation and growth rates is also daunting because of the presence of many simultaneous growth
processes and pathways.39 Furthermore, the application of computational methods such as
molecular dynamics (MD) simulations provides only limited information because of the short time
scales that are accessible.40
Recently, Smith et al. demonstrated a novel synthetic method that prevents agglomeration of COF
particles and instead results in a stable colloidal suspension.41 This approach enabled the temporal
resolution of the growth and nucleation processes and the synthesis of several single-crystalline
2D COF colloids.42 However, the underlying mechanism remains poorly understood, which stands
as a barrier to the general application of this approach. Here, we use kinetic Monte Carlo (KMC)
simulations to study the 2D COF nucleation and elongation rates as a function of initial monomer
5
concentration. Based on the simulation results, we are able to reach a quantitative understanding
of the nucleation and growth rates, which we further validate via in situ X-ray scattering
measurements. Overall our results allow us to rationalize why reducing the steady-state monomer
concentration suppresses nucleation and favors crystallite growth, the strategy that has been
successfully used in the fabrication of micron-sized 2D COF crystals.42 This kinetic model will
inspire other strategies to temporally resolve nucleation and growth. Similar control of initiation
and propagation in linear polymerization has revolutionized precision polymer synthesis, and its
further development in two-dimensional systems is key to fully realizing the potential of this
longstanding missing topology in macromolecular science.
Results and Discussion
To probe the general dynamics of 2D COF polymerization, we focus on the boronate ester-linked
COF-5, which is a frequently used model system in the study of 2D COFs.35,38,41-45 COF-5 is
formed from the condensation of 2,3,6,7,10,11-hexahydroxytriphenylene (HHTP) and 1,4-
phenylenebis(boronic acid) (PBBA), as shown in Scheme 1. We have previously developed a
kinetic Monte Carlo (KMC) model for 2D COF formation in solution, which makes use of a
combination of experimentally and computationally derived parameters.38 The rate parameters
related to bond formation, bond breakage, and van der Waals interactions involved in the COF-5
growth have been parameterized in our previous work and found to reproduce very well the
experimental production curves.38 Here, in order to fully quantify the nucleation and growth rates,
we carried out two sets of KMC simulations, based on modifications from our original KMC
model.
6
Scheme 1. The two-step growth process of micron-sized crystals obtained by slowly adding
monomers to COF nanoparticle seeds42.
The first set of simulations is designed to evaluate the nucleation rates. Here, the nuclei that have
grown beyond a threshold size of 50 monomers are removed from consideration, thereby
suppressing the impact of growth. The initial system contains 48,200 HHTP molecules and 72,300
PBBA molecules; the volume is varied in order to model different initial monomer concentrations.
These amounts were chosen to minimize any finite-size effects (see Figure S2 in the Supporting
Information, SI) while still allowing for practical simulation times. The simulations are stopped
when the nuclei that are generated correspond to 10% conversion of all monomers; in this way,
the monomer concentrations do not vary significantly from their initial conditions. The nucleation
rates are calculated through a linear fitting of nucleus production with time (additional
computational details can be found in the SI). For each of the initial conditions, 20 simulations
were run in order to reduce the statistical errors.
The second set of simulations focuses on the evaluation of growth rates by preventing additional
nucleation after the appearance of the first nuclei. To model reliably large crystals, the system size
is increased by a factor of ten so as to hold 482,000 HHTP molecules and 723,000 PBBA
molecules. The simulations are completed when the crystal has grown to a size of 100,000
7
monomer units (which corresponds to less than 10% of the total number of monomers). The in-
plane expansion/stacking rates are calculated by considering the evolution of the particle
diameter/height vs. time. Similar to the first set of simulations, in order to obtain different initial
monomer concentrations, the volume is varied while keeping the same number of monomers.
The results of the KMC simulations are shown in Figure 1. They establish the following
fundamental relationships among the nucleation rate (Jnuc), the in-plane (lateral) expansion rate
(vin-plane), and the stacking (vertical expansion) rate (vstacking), with respect to the monomer
concentration (CMonomer = CHHTP = 2/3 CPBBA) –note that, for the sake of simplicity, we neglect the
change in monomer concentration with respect to its initial value:
2
nuc nucleation monomerJ k C= (1)
,in plane growth in plane monomerv k C− −= (2)
,stacking growth stacking monomerv k C= (3)
By considering a series of simulated monomer concentrations, we obtain that the knucleation, kgrowth,in-
plane, and kgrowth,stacking rate coefficients have values of 0.024 L.mol-1.s-1, 1,418 nm.L.mol-1.s-1 and
599 nm.L.mol-1.s-1, respectively.
8
Figure 1. A) KMC-simulated nucleation rate as a function of monomer concentration. B) KMC-
simulated in-plane and stacking growth rates as a function of monomer concentration.
Our earlier KMC simulations showed that nucleation involves bond formation and stacking
between oligomers, which are both second-order in the reactants.38 At steady state, the nucleation
speed is expected to be governed by one of these two steps. Regardless of which step is the limiting
one, a second-order dependence is preserved, as shown by Eq. (1). The first-order dependence of
growth on monomer concentration can be rationalized by considering that, while any growth is
formally a second-order process, given that one reactant is fixed in the crystal, the overall dynamics
become pseudo first-order.
9
Recently, Smith et al. reported an original methodology that allowed them to study COF-5
particles as homogenous colloidal suspensions.41 These COF-5 nanoparticle solutions are much
more suitable for mechanistic investigations than conventional agglomerated particles, since the
heterogeneous processes of precipitation and aggregation are suppressed. Thus, we have measured
the growth rates of the nanocrystalline colloidal suspensions in order to characterize in depth the
mechanism of 2D COF growth.
To probe the rate dependence of COF-5 nucleation or growth, we turned to in situ
Small/Medium/Wide-Angle X-ray Synchrotron Scattering (SAXS/MAXS/WAXS)
measurements. The simultaneous collection of these data sets allowed us to monitor the size and
crystallinity of the particles in tandem, as the reaction proceeded. To probe the nucleation and
growth processes independently, it was necessary to choose conditions under which the processes
were temporally resolved. This was achieved by selecting monomer concentrations under which
one process dominated, as was seen for COF-5 in a recent report by Evans et al.42 To probe the
nucleation behavior of COF-5, we prepared solutions of HHTP and PBBA at concentrations of
over 1 mM, which were recently demonstrated to favor nucleation. After heating these solutions
to 70 °C, we continuously monitored the MAXS pattern during COF-5 nucleation (Figure 2A).
The integrated intensity of the <100> Bragg feature (~0.24 Å-1) as a function of time was extracted
from these data (Figure 2B, inset). Finally, linear fits of signal intensity versus time were extracted
to determine the rate of nucleation under these conditions. We observe a second-order relationship
between the initial monomer concentration and the growth rate of the <100> signal intensity
(Figure 2B). This observation is thus consistent with the results of the KMC simulations described
above and supports a second-order dependence of COF nucleation on monomer concentration.
10
Figure 2. A) MAXS traces as a function of time for selected data of B. B) Rate of intensity of the
<100> Bragg feature as a function of initial monomer concentration (inset: <100> intensity as a
function of time for selected monomer concentrations). C) MAXS traces as a function of time for
selected data of D. D) Rate of increase of intensity of the <100> Bragg feature as a function of
monomer equivalents added (inset: <100> intensity as a function of time for selected equivalents
added).
In order to probe the monomer dependence of growth, we sought to identify conditions under
which nucleation would be suppressed, yet growth would occur at measurable rates. We first
determined an upper concentration limit for growth experiments (0.7 mM) by allowing HHTP and
PBBA to react in the absence of preexisting seed particles. No change in signal was observed,
indicating that nucleation is suppressed at these time scales (Figure S9). It was also critical to
identify concentrations of COF seeds that were observable while sufficiently dilute to allow for
observable growth, which was ultimately optimized to an initial seed concentration. Mixtures of
COF-5 seed particles at a constant particle density but variable free monomer equivalents (moles
HHTP free monomer / moles of HHTP present in COF seeds) below this nucleation threshold were
then prepared and heated to 70 °C (see SI for more details). Additionally, a sample was prepared
11
by diluting the COF-5 seeds in the absence of additional monomers. Once again, the MAXS pattern
was monitored as a function of time (Figure 2C), with extracted <100> intensities (Figure 2D
inset) and rate of this intensity growth (Figure 2D) being determined as described for the
nucleation experiment. In this case, the <100> intensity growth rate is clearly linear with respect
to the monomer equivalents (and therefore the initial monomer concentration) being added. As
expected, under conditions where no additional monomers were added to the diluted COF seeds,
the initial seed particles remained unchanged (Figure S10), which confirms that any enhancement
in signal is not related to processes such as Ostwald ripening. Taken together, these results indicate
that growth is a first-order process with respect to monomer concentration when the particle
density of COF-5 seeds is held constant.
We stress that precise control of nucleation and growth rates is critical to fabricating high-quality
2D COF crystals. Based on the above results, it can be concluded that for COF-5 particles
nucleation is an inherently higher-order process than growth. As we elaborate on in the following
discussion, this difference in dynamics is the origin of the observation that nucleation and growth
can be temporally resolved by controlling the monomer concentration, with nucleation being
suppressed at lower monomer concentrations.42,46
We now turn to a discussion of the relative strengths of nucleation and growth rates. One way to
compare these rates is through monomer consumption. The monomer consumption rates for
nucleation (ωmonomer,nuc) and growth (ωmonomer,growth) of a crystal take on the following forms when
assuming a cylindrical shape and a constant diameter-to-height ratio:
2
, nucNmonomer nuc nuc monomerJ C = (4)
12
2
,monomer growth monomer
dVd C
dt = (5)
where Nnuc is the nucleus size, taken here as 50 monomer units; β, the number of monomer units
per unit volume in the crystal; V, the crystal volume; t, the time; and d, the crystal diameter.
If we consider the monomer consumption rates of a colloidal system consisting of uniform
microcrystals of diameter d, the steady-state nucleation and growth rates are sketched in Figure
3A. Therefore, the second- and first-order dependences of nucleation and growth rates on
monomer concentration mean that there always exists a threshold concentration below which
nucleation is slower than growth (Figure 3A). This analysis thus rationalizes the recent findings
that nucleation is suppressed when adding monomers slowly to the colloidal solution.42,46
Figure 3. A) Comparison of the monomer consumption dynamics of nucleation and growth. B)
Illustration of the critical monomer concentration and reaction time at different desired crystallite
expansion (l). The value of Cnuc,0 is assumed to be 3.3×10-5 mM.
13
Understanding the relative rates of competing processes allows for guided experimental design of
COF crystal growth by selecting a specific, steady-state monomer concentration. Higher monomer
concentrations will speed the growth of existing colloids, yet complicates the suppression of the
nucleation of new particles. Considering the interplay between nucleation and growth processes
are always at play, it is challenging to completely eliminate nucleation. However, with the
objective of suppressing nucleation, we define γ as the number of newly nucleated species divided
by the initial seed crystals (it can be understood as a proportionality factor between the occurrence
of new crystals with respect to the existing ones) during a growth process in which the diameter
of the existing crystals is extended by l. There exists a critical monomer concentration (C*) that
represents the upper theoretical limit for a specific combination of γ and l. By this definition, the
monomer concentration in the reaction needs to be lower than C* to achieve the desired crystal
expansion while suppressing nucleation. Monomer concentrations above C* lead to more frequent
nucleation (increased γ) and smaller average crystal sizes. We find that C* depends on the initial
nuclei concentration (Cnuc,0) and can be expressed as:
,0 ,* nuc growth in plane
nucleation
C kC
lk
−= (6)
The monomer consumption rate ωconsump (from both nucleation and growth) gives information
about the monomer amount that should be added to maintain a constant monomer concentration at
Cmonomer. When assuming the crystals to be cylindrical and under the condition that γ≪1, we find
that ωconsump is expressed as:
( )2, 2
, 0 nuc
3N
4
growth in plane nuclei monomer
consump growth in plane monomer nucleation monomer
k C Ck C t d k C
−
−= + + (7)
14
where α is the diameter-to-height ratio of the crystal and d0, the initial diameter of the crystals. It
emerges from Eq. (7) that ωconsump is not constant but instead increases quadratically with time.
For practical 2D COF growth, the reaction time is also an important factor to be considered. The
use of too low a monomer concentration should be avoided as it results in undesirably slow growth.
The shortest time for crystal growth is achieved when using Cmonomer = C*. In this case, the time
required for the desired crystal expansion l is calculated to be:
,
2
* 2
, ,0growth in plane
nucleation
growth in plane nuclei
k ll
k C k C
−−
= = (8)
The quadratic dependence of time on desired growth coupled with the inverse dependence on γ
means that growing ever larger crystals by slow monomer addition is likely to face practical
challenges. Nevertheless, this analysis provides a theoretical reference for the monomer
concentration to use during 2D COF fabrication, the amount of monomers to add, and the estimated
time of crystal growth. This theoretical foundation is critical to the rational control over 2D COF
synthetic conditions, and therefore temporal resolution of different microscale processes relevant
to COF growth. Furthermore, the model we have developed for COF growth suggests that
mechanisms other than direct condensation (e.g., formal transimination 47 and monomer exchange
reactions 48) need to be explored to determine whether growth can be favored over nucleation,
which could yield 2D COF single crystals under more practical time scales. In addition, we
anticipate that combinations of strategies will likely be more effective than individual approaches.
15
Conclusions
The combination of kinetic Monte Carlo simulations and in situ X-ray scattering measurements
has allowed us to develop a quantitative analysis of the nucleation and growth rates of 2D COFs,
as represented by COF-5. The nucleation and growth processes have second-order and first-order
dependences on monomer concentration, respectively. These different reaction orders explain why
growth can be suppressed by controlling the speed of monomer addition. Based on this
experimentally validated model, we were able to define a critical monomer concentration C*,
which provides insight into the optimal monomer concentration to use for 2D COF growth via
slow addition of monomers to 2D COF colloidal seeds.
This understanding allows us to express several rules for the optimization of 2D COF synthesis.
First, we find that C* is proportional to the occurrence of new crystals with respect to the existing
ones, while inversely proportional to the desired amount of expansion of the seed. Therefore, lower
monomer concentrations should be selected when targeting fewer and larger crystallite domains.
Secondly, C* is linearly dependent on the number of nuclei in solution. This means that higher
monomer concentrations may be used in solutions where the number of nuclei is also larger.
Therefore, in solutions with more COF nanoparticles, the critical monomer concentration is
increased as well as the rate of monomer consumption directed toward crystal growth. However,
this growth will be shared among the larger number of crystallite seeds, making this strategy
ineffective for increasing the average crystallite domain size more quickly.
The fundamental connection between C* and the competition between nucleation and growth
means that the reaction time has a quadratic dependence on the desired crystal expansion. While
no theoretical upper limit exists in terms of a maximum COF size, the nonlinear dependence on
16
time suggests that growing large crystals can rapidly become impractical. For example, as
illustrated by Eq. (8), growing 1-mm long crystals for a value of γ (the proportionality factor
between the occurrence of new crystals with respect to the existing ones) equal to 0.01, can take
over 1,000,000 years. Thus, growing millimeter-sized 2D COF crystals from solution using a
direct-condensation, seeded-growth approach remains challenging. However, this limitation is
associated with the specific strategy of limiting nucleation only by maintaining a low monomer
concentration. Other methods to limit nucleation, such as chemically increasing the
transesterification rate associated with COF formation, might provide access to high-quality COFs
under different kinetic regimes. New synthetic strategies are called for to achieve these goals.
Finally, we note that 2D COF formation and growth can rely on different mechanisms. Here, we
have used COF-5 as a representative example and the conclusions thus likely apply to the series
of boronate ester-linked COFs. It will be of interest to determine whether the set of empirical
equations developed here can be extended to other types of 2D COFs for which different growth
mechanisms have been proposed. Such mechanistic investigations of chemically distinct systems
are part of our ongoing collaborative efforts.
ASSOCIATED CONTENT
Supporting Information. Additional details on KMC simulations, the calculations of nucleation
and growth rates and in situ XRD measurements. This material is available free of charge via the
Internet at http://pubs.acs.org.
17
AUTHOR INFORMATION
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENTS
This work was supported by the Army Research Office, under the Multidisciplinary University
Research Initiative (MURI) program, Award No. W911NF-15-1-0447, and under Grant No.
W911NF-17-1-0339. A. M. E. (DGE-1324585), M.J.S. (DGE-1842165), and I.C. (DGE-1842165)
are supported by the National Science Foundation Graduate Research Fellowship. I.C. is partially
supported by the Ryan Fellowship and the International Institute for Nanotechnology. Parts of this
work were performed at the DuPont-Northwestern-Dow Collaborative Access Team (DND-CAT)
located at Sector 5 of the Advanced Photon Source (APS); DND-CAT is supported by
Northwestern University, E.I. DuPont de Nemours & Co., and the Dow Chemical Company. This
research used resources of the Advanced Photon Source and Center for Nanoscale Materials, both
U.S. Department of Energy (DOE) Office of Science User Facilities operated for the DOE Office
of Science by Argonne National Laboratory under Contract No. DE-AC0206CH11357.
REFERENCES
(1) Sakamoto, J.; van Heijst, J.; Lukin, O.; Schlüter, A. D., Two-Dimensional Polymers: Just
a Dream of Synthetic Chemists?, Angew. Chem., Int. Ed. 2009, 48, 1030.
(2) Cote, A. P.; Benin, A. I.; Ockwig, N. W.; O'Keeffe, M.; Matzger, A. J.; Yaghi, O. M.,
Porous, Crystalline, Covalent Organic Frameworks, Science 2005, 310, 1166.
(3) Feng, X.; Ding, X. S.; Jiang, D. L., Covalent Organic Frameworks, Chem. Soc. Rev. 2012,
41, 6010.
(4) Wu, D. C.; Xu, F.; Sun, B.; Fu, R. W.; He, H. K.; Matyjaszewski, K., Design and
Preparation of Porous Polymers, Chem. Rev. 2012, 112, 3959.
18
(5) Colson, J. W.; Dichtel, W. R., Rationally Synthesized Two-Dimensional Polymers, Nature
Chem. 2013, 5, 453.
(6) Ding, S. Y.; Wang, W., Covalent Organic Frameworks (COFs): From Design to
Applications, Chem. Soc. Rev. 2013, 42, 548.
(7) Bertrand, G. H. V.; Michaelis, V. K.; Ong, T.-C.; Griffin, R. G.; Dincă, M., Thiophene-
Based Covalent Organic Frameworks, Proc. Natl. Acad. Sci. 2013, 110, 4923.
(8) Xu, H.; Gao, J.; Jiang, D., Stable, Crystalline, Porous, Covalent Organic Frameworks as a
Platform for Chiral Organocatalysts, Nature Chem. 2015, 7, 905.
(9) Waller, P. J.; Gandara, F.; Yaghi, O. M., Chemistry of Covalent Organic Frameworks, Acc.
Chem. Res. 2015, 48, 3053.
(10) Huang, N.; Wang, P.; Jiang, D., Covalent Organic Frameworks: A Materials Platform for
Structural and Functional Designs, Nat. Rev. Mater. 2016, 1, 16068.
(11) Ascherl, L.; Sick, T.; Margraf, J. T.; Lapidus, S. H.; Calik, M.; Hettstedt, C.; Karaghiosoff,
K.; Döblinger, M.; Clark, T.; Chapman, K. W.; Auras, F.; Bein, T., Molecular Docking Sites
Designed for the Generation of Highly Crystalline Covalent Organic Frameworks, Nature Chem.
2016, 8, 310.
(12) Beuerle, F.; Gole, B., Covalent Organic Frameworks and Cage Compounds: Design and
Applications of Polymeric and Discrete Organic Scaffolds, Angew. Chem., Int. Ed. 2018, 57, 4850.
(13) Lohse, M. S.; Bein, T., Covalent Organic Frameworks: Structures, Synthesis, and
Applications, Adv. Funct. Mater. 2018, 28, 1705553.
(14) Bisbey, R. P.; Dichtel, W. R., Covalent organic frameworks as a platform for
multidimensional polymerization, Acs Central Science 2017, 3, 533.
(15) DeBlase, C. R.; Silberstein, K. E.; Truong, T. T.; Abruna, H. D.; Dichtel, W. R., β-
Ketoenamine-linked covalent organic frameworks capable of pseudocapacitive energy storage, J.
Am. Chem. Soc. 2013, 135, 16821.
(16) Sun, B.; Zhu, C. H.; Liu, Y.; Wang, C.; Wan, L. J.; Wang, D., Oriented Covalent Organic
Framework Film on Graphene for Robust Ambipolar Vertical Organic Field-Effect Transistor,
Chem. Mater. 2017, 29, 4367.
(17) Miao, J.; Xu, Z.; Li, Q.; Bowman, A.; Zhang, S.; Hu, W.; Zhou, Z.; Wang, C., Vertically
Stacked and Self-Encapsulated van der Waals Heterojunction Diodes Using Two-Dimensional
Layered Semiconductors, ACS Nano 2017, 11, 10472.
(18) Dey, K.; Pal, M.; Rout, K. C.; Kunjattu H, S.; Das, A.; Mukherjee, R.; Kharul, U. K.;
Banerjee, R., Selective Molecular Separation by Interfacially Crystallized Covalent Organic
Framework Thin Films, J. Am. Chem. Soc. 2017, 139, 13083.
(19) Medina, D. D.; Sick, T.; Bein, T., Photoactive and Conducting Covalent Organic
Frameworks, Adv. Energy. Mater. 2017, 7, 1700387.
(20) Xiang, Z. H.; Cao, D. P.; Dai, L. M., Well-defined two dimensional covalent organic
polymers: rational design, controlled syntheses, and potential applications, Polym. Chem. 2015, 6,
1896.
(21) Pyles, D. A.; Crowe, J. W.; Baldwin, L. A.; McGrier, P. L., Synthesis of Benzobisoxazole-
Linked Two-Dimensional Covalent Organic Frameworks and Their Carbon Dioxide Capture
Properties, ACS Macro Lett. 2016, 5, 1055.
(22) Xu, Q.; Tao, S.; Jiang, Q.; Jiang, D., Ion Conduction in Polyelectrolyte Covalent Organic
Frameworks, J. Am. Chem. Soc. 2018, 140, 7429.
(23) Zhao, F.; Liu, H.; Mathe, S.; Dong, A.; Zhang, J., Covalent Organic Frameworks: From
Materials Design to Biomedical Application, Nanomaterials 2018, 8, 15.
19
(24) Zhao, W.; Xia, L.; Liu, X., Covalent organic frameworks (COFs): perspectives of
industrialization, CrystEngComm 2018, 20, 1613.
(25) Wan, S.; Guo, J.; Kim, J.; Ihee, H.; Jiang, D., A Belt-Shaped, Blue Luminescent, and
Semiconducting Covalent Organic Framework, Angew. Chem., Int. Ed. 2008, 47, 8826.
(26) Wan, S.; Gándara, F.; Asano, A.; Furukawa, H.; Saeki, A.; Dey, S. K.; Liao, L.; Ambrogio,
M. W.; Botros, Y. Y.; Duan, X.; Seki, S.; Stoddart, J. F.; Yaghi, O. M., Covalent Organic
Frameworks with High Charge Carrier Mobility, Chem. Mater. 2011, 23, 4094.
(27) Duhović, S.; Dincă, M., Synthesis and Electrical Properties of Covalent Organic
Frameworks with Heavy Chalcogens, Chem. Mater. 2015, 27, 5487.
(28) Crowe, J. W.; Baldwin, L. A.; McGrier, P. L., Luminescent Covalent Organic Frameworks
Containing a Homogeneous and Heterogeneous Distribution of Dehydrobenzoannulene Vertex
Units, J. Am. Chem. Soc. 2016, 138, 10120.
(29) Jin, E.; Asada, M.; Xu, Q.; Dalapati, S.; Addicoat, M. A.; Brady, M. A.; Xu, H.; Nakamura,
T.; Heine, T.; Chen, Q.; Jiang, D., Two-Dimensional sp2 Carbon–Conjugated Covalent Organic
Frameworks, Science 2017, 357, 673.
(30) Haldar, S.; Chakraborty, D.; Roy, B.; Banappanavar, G.; Rinku, K.; Mullangi, D.; Hazra,
P.; Kabra, D.; Vaidhyanathan, R., Anthracene-Resorcinol Derived Covalent Organic Framework
as Flexible White Light Emitter, J. Am. Chem. Soc. 2018, 140, 13367.
(31) Albacete, P.; Martínez, J. I.; Li, X.; López-Moreno, A.; Mena-Hernando, S. a.; Platero-
Prats, A. E.; Montoro, C.; Loh, K. P.; Pérez, E. M.; Zamora, F., Layer-Stacking-Driven
Fluorescence in a Two-Dimensional Imine-Linked Covalent Organic Framework, J. Am. Chem.
Soc. 2018, 140, 12922.
(32) Wan, S.; Guo, J.; Kim, J.; Ihee, H.; Jiang, D. L., A Belt-Shaped, Blue Luminescent, and
Semiconducting Covalent Organic Framework, Angew. Chem., Int. Ed. Engl. 2008, 47, 8826.
(33) Sick, T.; Hufnagel, A. G.; Kampmann, J.; Kondofersky, I.; Calik, M.; Rotter, J. M.; Evans,
A.; Döblinger, M.; Herbert, S.; Peters, K.; Böhm, D.; Knochel, P.; Medina, D. D.; Fattakhova-
Rohlfing, D.; Bein, T., Oriented Films of Conjugated 2D Covalent Organic Frameworks as
Photocathodes for Water Splitting, J. Am. Chem. Soc. 2018, 140, 2085.
(34) Gonçalves, R. S. B.; de Oliveira, A. B. V.; Sindra, H. C.; Archanjo, B. S.; Mendoza, M.
E.; Carneiro, L. S. A.; Buarque, C. D.; Esteves, P. M., Heterogeneous Catalysis by Covalent
Organic Frameworks (COF): Pd(OAc)2@COF-300 in Cross-Coupling Reactions, ChemCatChem
2016, 8, 743.
(35) Smith, B. J.; Dichtel, W. R., Mechanistic Studies of Two-Dimensional Covalent Organic
Frameworks Rapidly Polymerized from Initially Homogenous Conditions, J. Am. Chem. Soc.
2014, 136, 8783.
(36) Smith, B. J.; Hwang, N.; Chavez, A. D.; Novotney, J. L.; Dichtel, W. R., Growth Rates and
Water Stability of 2D Boronate Ester Covalent Organic Frameworks, Chem. Commun. 2015, 51,
7532.
(37) Medina, D. D.; Rotter, J. M.; Hu, Y.; Dogru, M.; Werner, V.; Auras, F.; Markiewicz, J. T.;
Knochel, P.; Bein, T., Room Temperature Synthesis of Covalent–Organic Framework Films
through Vapor-Assisted Conversion, J. Am. Chem. Soc. 2015, 137, 1016.
(38) Li, H. Y.; Chavez, A. D.; Li, H. F.; Li, H.; Dichtel, W. R.; Brédas, J.-L., Nucleation and
Growth of Covalent Organic Frameworks from Solution: The Example of COF-5, J. Am. Chem.
Soc. 2017, 139, 16310.
(39) De Yoreo, J. J.; Gilbert, P. U. P. A.; Sommerdijk, N. A. J. M.; Penn, R. L.; Whitelam, S.;
Joester, D.; Zhang, H. Z.; Rimer, J. D.; Navrotsky, A.; Banfield, J. F.; Wallace, A. F.; Michel, F.
20
M.; Meldrum, F. C.; Colfen, H.; Dove, P. M., Crystallization by particle attachment in synthetic,
biogenic, and geologic environments, Science 2015, 349, aaa6760.
(40) Sosso, G. C.; Chen, J.; Cox, S. J.; Fitzner, M.; Pedevilla, P.; Zen, A.; Michaelides, A.,
Crystal nucleation in liquids: Open questions and future challenges in molecular dynamics
simulations, Chem. Rev. 2016, 116, 7078.
(41) Smith, B. J.; Parent, L. R.; Overholts, A. C.; Beaucage, P. A.; Bisbey, R. P.; Chavez, A.
D.; Hwang, N.; Park, C.; Evans, A. M.; Gianneschi, N. C.; Dichtel, W. R., Colloidal Covalent
Organic Frameworks, ACS Central Science 2017, 3, 58.
(42) Evans, A. M.; Parent, L. R.; Flanders, N. C.; Bisbey, R. P.; Vitaku, E.; Kirschner, M. S.;
Schaller, R. D.; Chen, L. X.; Gianneschi, N. C.; Dichtel, W. R., Seeded Growth of Single-Crystal
Two-Dimensional Covalent Organic Frameworks, Science 2018, 361, 52.
(43) Li, H. F.; Li, H. Y.; Dai, Q. Q.; Li, H.; Brédas, J. L., Hydrolytic Stability of Boronate
Ester‐Linked Covalent Organic Frameworks, Adv. Theory Simul. 2018, 1, 1700015.
(44) Li, H. Y.; Brédas, J.-L., Large Out-of-Plane Deformations of Two-Dimensional Covalent
Organic Framework (COF) Sheets, J. Phys. Chem. Lett. 2018, 9, 4215.
(45) Li, H. Y.; Brédas, J.-L., Nanoscrolls Formed from Two-Dimensional Covalent Organic
Frameworks, Chem. Mater. 2019, 31, 3265.
(46) Li, Rebecca L.; Flanders, N. C.; Evans, A. M.; Ji, W.; Castano, I.; Chen, L. X.; Gianneschi,
N. C.; Dichtel, W. R., Controlled growth of imine-linked two-dimensional covalent organic
framework nanoparticles, Chem. Sci. 2019, 10, 3796.
(47) Vitaku, E.; Dichtel, W. R., Synthesis of 2D Imine-Linked Covalent Organic Frameworks
through Formal Transimination Reactions, J. Am. Chem. Soc. 2017.
(48) Daugherty, M. C.; Vitaku, E.; Li, R. L.; Evans, A. M.; Chavez, A. D.; Dichtel, W. R.,
Improved synthesis of β-ketoenamine-linked covalent organic frameworks via monomer exchange
reactions, Chem. Commun. 2019, 55, 2680.
download fileview on ChemRxivCOFNucleationGrowth_062119_ChemRxiv.pdf (0.90 MiB)
S1
Nucleation-Elongation Dynamics of Two-Dimensional
Covalent Organic Frameworks
Supplementary Information
Haoyuan Lia§, Austin M. Evansb§, Ioannina Castanob, Michael J. Straussb, William R. Dichtelb*
and Jean-Luc Bredasa*
aSchool of Chemistry and Biochemistry, Center for Organic Photonics and Electronics (COPE),
Georgia Institute of Technology, Atlanta, Georgia 30332-0400 bDepartment of Chemistry, Northwestern University, Evanston, IL 60208, USA.
Corresponding Authors
Table of Contents
I. Simulation Results ............................................................................ S2
II. Materials and Instrumentation ...................................................... S7
III. Synthesis of COF-5 Seeds ............................................................. S8
IV. Experimental in situ Scattering Details ....................................... S9
V. X-Ray Scattering Control Experiments ...................................... S12
VI. In Situ X-Ray Scattering Nucleation Results ............................ S14
VII. In Situ X-Ray Scattering Growth Results ................................ S20
S2
I. Simulation Results
The parameters used in the KMC simulations have been documented in our previous work.1 In the
present KMC simulations, nuclei have been identified by setting a size threshold of 50 monomer
units, which we choose to be slightly larger than the value of 40 monomer units used in our
previous study 1 in order to ensure better identification. Figure S1 shows a typical nucleus
production curve. For each simulation, a linear fitting is performed. The nucleation rate
corresponds to the fitted slope.
Figure S1. Representative Linear Fitting Used in the KMC Simulations: Illustration of the
calculation of the nucleation rates.
S3
For nucleation simulations, the total number of monomers is 120,500. We have checked that the
error coming from the finite sizes of the simulation system is not significant, as shown in Figure
S2.
Figure S2. Simulation Results for Different System Sizes in the KMC Simulations. N is the
total number of monomers. The concentrations of monomers are the same across all simulations.
The lateral expansion is found to be approximately linear with time. Figures S3 and S4 show a
typical COF expansion in the course of the simulated growth and the top view of the final crystal,
respectively. Linear fittings are performed on the crystal expansion results, see Figure S5. The
lateral expansion rate thus corresponds to the fitted slope.
S4
Figure S3. Example of KMC Growth Result. Illustration of the evolution of average diameter
and height as a function of time.
S5
Figure S4. COF-5 particle simulation. Illustration of the final structure of the COF-5 crystal (top
view and side view) from the KMC simulation.
S6
Figure S5. Fitting of the Lateral Expansion Data. Lateral expansion as a function of time and
linear fittings.
S7
II. Materials and Instrumentation
Materials
Acetonitrile, 1-4 Dioxane, and Mesitylene were purchased from Fisher Scientific.
Phenylbisboronic acid (PBBA) was purchased from Sigma-Aldrich and hexahydroxytriphenylene
(HHTP) was purchased from TCI America. All chemicals were used without further purification.
Instrumentation
Sonication. Sonication was performed with a Branson 3510 ultrasonic cleaner with a power
output of 100W and a frequency of 42 kHz.
In Situ X-ray Scattering: Small/Medium/Wide-Angle X-ray Scattering (SAXS/MAXS/WAXS)
data were collected simultaneously at sector 5-ID-D of the Advanced Photon Source at Argonne
National Laboratory. A beam energy of 17.0 KeV was used for these experiments. Patterns were
collected with single 10 s frames on a series of 3 Pilatus 2D detectors which were then radially
integrated. All samples were conducted in 1.5 mm borosilicate capillaries with a wall thickness of
0.01 mm available from Charles Supper Scientific. Patterns were assigned to be time = 0 seconds
when the heater reached its steady state temperature of 70 °C (approximately 45 seconds). Patterns
were baselined and then background subtracted from the starting time pattern to produce corrected
patterns. The <100> Bragg feature of COF-5 (~0.24 Å-1) was then integrated and plotted against
the time of each pattern. Finally, the rate of increase in <100> signal intensity was determined by
fitting the slope of the respective <100> intensity versus time for each experiment.
S8
III. Synthesis of COF-5 Seeds
Figure S6. Synthesis Procedure of COF-5 Colloidal Seeds.
Hexahydroxytriphenylene (HHTP) and phenylbisboronicacid (PBBA) were added to a 20 mL
mixture of CH3CN:dioxane:mesitylene (80:16:4 vol%) at a concentration of 2 mM HHTP and 3
mM PBBA. This solution was sonicated until full dissolution of the monomers was observed. This
solution was then heated in a scintillation vial overnight at 70 °C after which time the solution
became opaque indicating the successful synthesis of COF-5 seed particles. These seeds were then
used as prepared for COF growth experiments detailed below. This method is explained in further
detail in Ref. 2
S9
IV. Experimental in situ Scattering Details
Figure S7. Nucleation Experiment. Schematic of how the nucleation experiments were
conducted.
Nucleation Experiment: Hexahydroxytriphenylene (HHTP) and phenylbisboronicacid (PBBA)
were added to a 20 mL mixture of CH3CN:dioxane:mesitylene (80:16:4 vol%) in a 1:1.5 molar
ratio at a concentration of double that specified in the experiment. For example, if the nucleation
experiment was conducted at 2 mM, an HHTP solution of 4 mM and a PBBA solution of 6 mM
would have been made independently. This is to prevent reaction before the mixing of the two
monomers. These solutions were sonicated until full solvation of the monomers. The two monomer
solutions were then mixed in a dram vial in a 1:1 volume ratio (final volume 1 mL) and were then
diluted as specified to produce the specified monomer concentration. These solutions were then
quickly shaken to ensure mixing. Approximately 100 µL of these solutions were then added to a
borosilicate capillary and sealed using capillary sealing putty. These capillaries were immediately
installed onto a multicapillary heating stage available at Sector 5-ID-D of the Advanced Photon
S10
Source (APS) Argonne National Lab (ANL). This stage was then heated to 70 °C (< 1 min heating
time) and allowed to react while continuous SAXS/MAXS/WAXS patterns were taken.
Figure S8. Growth Experiment. Schematic of how the growth experiments were conducted.
Growth Experiment: Hexahydroxytriphenylene (HHTP) and phenylbisboronicacid (PBBA)
were added to a separate 20 mL mixture of CH3CN:dioxane:mesitylene (80:16:4 vol%) at a
concentration of 1 mM HHTP and 1.5 mM PBBA. These solutions were sonicated until full
solvation of the monomers. The two monomer solutions were then mixed in a dram vial in a 1:1
volume ratio (final volume 2 mL) and quickly shaken. Dilutions were then performed in dram vials
to produce different concentrations of monomer solutions with the same volume. The highest
monomer concentration (0.48 mM) was chosen to temporally resolve nucleation and growth. The
same amount of COF-5 seeds (COF-5 seeds diluted by a factor of 20, resulting in particle solution
containing 0.1 mM HHTP and 0.15 mM PBBA) were then added to all reactions to keep the density
of seed particles identical. To ensure that Ostwald ripening or depolymerization were not relevant
processes under these conditions, we also diluted our COF-5 seed particles and performed the
S11
experiment without the presence of additional monomers. Precise amounts of the dilution
experiment are recorded below.
Table S1. Dilution Calculations. Calculations of how dilutions were conducted for COF-5
growth experiments.
Approximately 100 µL of these solutions were then added to a borosilicate capillary and sealed
using capillary sealing putty. These capillaries were immediately installed onto a multicapillary
heating stage available at Sector 5-ID-D of the Advanced Photon Source (APS) Argonne National
Lab (ANL). This stage was then heated to 70 °C and allowed to react while continuous
SAXS/MAXS/WAXS patterns were taken.
Seed Stock
(mL)
Blank Solvent
(mL)
Monomer Amount
(mL)
Resultant Monomer
Concentration (mM)
0.1 0 1.9 0.48
0.1 0.4 1.5 0.38
0.1 0.6 1.3 0.33
0.1 0.8 1.1 0.28
0.1 1 0.9 0.23
0.1 1.2 0.7 0.18
0.1 1.9 0 0
S12
V. X-Ray Scattering Control Experiments
Figure S9. Nucleation Control. In situ X-ray scattering of 0.7 mM HHTP 1.1 mM PBBA solution
over the course of 1 hour.
The background subtracted MAXS data shown in the bottom right corner demonstrates that under
conditions where monomer concentration is sufficiently small results in no <100> diffraction
feature and thus no substantial COF nucleation on the time-scales studied here. The background
subtracted SAXS data shown in the bottom left corner shows that species of small sizes are formed
over the course of this reaction. This validates previous understanding of COF nucleation where
small oligomers are formed but never “nucleate” unless they reach a critical concentration.
S13
Figure S10. Growth Control. In situ X-ray scattering of COF-5 seeds heated in the presence of
no additional monomer.
The MAXS data shown in the top right corner demonstrates that under conditions where no
additional monomer is added no discernable difference in the <100> diffraction feature is noted.
The background subtracted SAXS data shown in the bottom left corner shows that there is no
discernable size change of the existing nucleation after the course of one hour. These experiments
suggest that the increase in intensity seen under monomer addition conditions are due to growth
and not processes such as Ostwald ripening. This is consistent with observations made previously.2
S14
VI. In Situ X-Ray Scattering Nucleation Results
All experiments were conducted over the course of approximately one hour. Conditions of each
experiment are given in their respective figure headings. In these figures [Monomer] = [HHTP] =
2/3*[PBBA] is the concentration of monomers present at the beginning of the experiment.
Figure S11. Nucleation Experiment [Monomer] = 0.9 mM (measurement 1).
S15
Figure S12. Nucleation Experiment [Monomer] = 0.9 mM (measurement 2).
Figure S13. Nucleation Experiment [Monomer] = 1.1 mM (measurement 1).
S16
Figure S14. Nucleation Experiment [Monomer] = 1.1 mM (measurement 2).
Figure S15. Nucleation Experiment [Monomer] = 1.3 mM (measurement 1).
S17
Figure S16. Nucleation Experiment [Monomer] = 1.3 mM (measurement 2).
Figure S17. Nucleation Experiment [Monomer] = 1.5 mM.
S18
Figure S18. Nucleation Experiment [Monomer] = 1.8 mM.
Figure S19. Nucleation Experiment [Monomer] = 2.3 mM.
S19
Figure S20. Nucleation Experiment [Monomer] = 3.0 mM.
S20
VII. In Situ X-Ray Scattering Growth Results
All experiments were conducted over the course of approximately one hour. Conditions of each
experiment are given in their respective figure headings. In these figures [Monomer] = [HHTP] =
2/3*[PBBA] is related to the amount of monomer added to a consistent amount and concentration
of preexisting COF-5 seeds.
Figure S21. Growth Experiment [Monomer] = 0.18 mM.
S21
Figure S22. Growth Experiment [Monomer] = 0.23 mM.
Figure S23. Growth Experiment [Monomer] = 0.28 mM.
S22
Figure S24. Growth Experiment [Monomer] = 0.33 mM.
Figure S25. Growth Experiment [Monomer] = 0.38 mM.
S23
Figure S26. Growth Experiment [Monomer] = 0.48 mM.
References
(1) Li, H. Y.; Chavez, A. D.; Li, H. F.; Li, H.; Dichtel, W. R.; Brédas, J.-L., Nucleation and
Growth of Covalent Organic Frameworks from Solution: The Example of COF-5, J. Am. Chem.
Soc. 2017, 139, 16310.
(2) Evans, A. M.; Parent, L. R.; Flanders, N. C.; Bisbey, R. P.; Vitaku, E.; Kirschner, M. S.;
Schaller, R. D.; Chen, L. X.; Gianneschi, N. C.; Dichtel, W. R., Seeded Growth of Single-Crystal
Two-Dimensional Covalent Organic Frameworks, Science 2018, 361, 52.
download fileview on ChemRxivCOFNucleationGrowth-SI-062119_ChemRxiv.pdf (3.59 MiB)