Radical-mediated step-growth: Preparation of hybrid polymer monolithic columns with fine control of nanostructural and chromatographic characteristics

Abstract

The currently most successful type of porous polymer monoliths utilized in chromatography is prepared by free-radical cross-linking (co)polymerization in porogenic solvents and a single-step molding process. Though such types of materials are well-recognized in the scientific community, they suffer from their multi-scale heterogeneity originating from the nanoscale through to their microscale and ultimately limited performance on their macroscale. This is in particular true when estimating their performance under equilibrium (i.e. isocratic) elution conditions for retained compounds.

Full text

Journal
of
Chromatography
A,
1412
(2015)
112–125
Contents
lists
available
at
ScienceDirect
Journal
of
Chromatography
A
jo
ur
nal
ho
me
pag
e:
www.elsevier.com/locate/chroma
Radical-mediated
step-growth:
Preparation
of
hybrid
polymer
monolithic
columns
with
fine
control
of
nanostructural
and
chromatographic
characteristics
Filipa
Alves,
Ivo
Nischang∗,1
Institute
of
Polymer
Chemistry,
Johannes
Kepler
University
Linz,
A-4060
Leonding,
Austria
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
24
June
2015
Received
in
revised
form
5
August
2015
Accepted
10
August
2015
Available
online
13
August
2015
Keywords:
Adsorption
Click
chemistry
Mass
transfer
Network
Ideality
Step-growth
Partition
a
b
s
t
r
a
c
t
The
currently
most
successful
type
of
porous
polymer
monoliths
utilized
in
chromatography
is
prepared
by
free-radical
cross-linking
(co)polymerization
in
porogenic
solvents
and
a
single-step
molding
process.
Though
such
types
of
materials
are
well-recognized
in
the
scientific
community,
they
suffer
from
their
multi-scale
heterogeneity
originating
from
the
nanoscale
through
to
their
microscale
and
ultimately
limited
performance
on
their
macroscale.
This
is
in
particular
true
when
estimating
their
performance
under
equilibrium
(i.e.
isocratic)
elution
conditions
for
retained
compounds.
In
this
contribution,
we
study
a
new
concept
in
the
preparation
of
porous
monolithic
hybrid
materials
based
on
polyhedral
oligomeric
vinylsilsesquioxanes
which
undergo
radical
mediated
step-growth
cross-
linking
with
thiol-linkers.
Fundamental
characterization
of
this
new
entry
of
materials
is
performed
via
a
variety
of
characterization
approaches
including
infrared
and
Raman
spectroscopies,
thermogravimet-
ric
analysis,
gel
fraction,
dry-state
surface
area
analysis,
and
visualization
of
the
capillary-scale
porous
structure
by
scanning
electron
microscopy.
This
characterization
identifies
that
a
rational
choice
of
exper-
imental
conditions
in
monolith
preparation
leads
to
destined
and
desirable
materials’
properties,
in
particular
with
experimentally
accessible
near-ideal
nanoscale
network
structures.
With
the
obtained
structural
informations
at
hand,
we
finally
evidence
the
monoliths’
tailored
chromatographic
perfor-
mance
by
isocratic
elution
experiments
of
structurally
similar
small
molecules
under
reversed-phase
type
of
chromatographic
conditions.
This
validates
the
fundamental
origin
for
an
improved
performance
of
these
types
of
monolithic
materials
under
solvated
conditions
that
has
its
foundation
established
in
the
creation
of
near-ideal
nanoscale
networks
of
material.
This
identified
ideality
is
manifested
in
an
enhanced
and
almost
retention-insensitive
performance
in
liquid
chromatographic
separations
of
small
molecules
across
wide
ranges
of
retention
factors
over
at
least
two
orders
of
magnitude
and
wide
ranges
of
mobile
phase
compositions.
Such
experimental
observation
is
explained
by
a
more
homogeneous
ener-
getic
distribution
of
partition
and
adsorption
sites.
A
reference
analysis
of
normalized
plate
height
data
at
varied
retention
was
performed
and
set
in
context
with
data
of
state-of-the-art
silica-
and
polymer-based
monoliths.
This
analysis
clearly
identifies
the
present
materials
to
display
performance
behavior
clearly
located
in
the
domain
of
derivatized
silica-based
monoliths.
©
2015
The
Authors.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY
license
(http://creativecommons.org/licenses/by/4.0/).
1.
Introduction
High
performance
chromatographic
separations
are
typically
performed
using
packed
beds
of
(porous)
adsorbent
particles
with
reversed-phase
liquid
chromatography
being
the
major
workhorse
in
related
industries
and
academic
studies.
Despite
its
success,
∗Corresponding
author.
Tel.:
+43
732
2468
7118;
fax:
+43
732
2468
7105.
E-mail
address:
ivo.nischang@jku.at
(I.
Nischang).
1Present
address:
Institute
of
Polymer
Science,
Johannes
Kepler
University
Linz,
Altenberger
Strasse
69,
A-4040
Linz,
Austria.
other
modes
of
chromatography
are
gaining
in
importance
as
well,
the
most
prominent
of
which
is
hydrophilic
interaction
liquid
chro-
matography
[1,2].
Overarching,
chromatography
desires
separation
of
complex
mixtures
of
compounds.
Fundamentally,
in
terms
of
mass
transport
in
both
reversed
phase
and
hydrophilic
interaction
chromato-
graphic
modes,
the
mechanisms
of
retention
can
be
described
by
both
partition
as
well
as
adsorption,
respectively
their
interplay
[3–5].
While
the
majority
of
stationary
phases
utilized
in
chromatog-
raphy
are
based
on
packed
beds,
the
paradigm
in
stationary
phase
design
was
altered
at
the
end
of
the
last
century
with
two
major
http://dx.doi.org/10.1016/j.chroma.2015.08.019
0021-9673/©
2015
The
Authors.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY
license
(http://creativecommons.org/licenses/by/4.0/).

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
113
breeds
of
monolithic
columns
entering
the
chromatographic
arena.
These
were
polymer
monoliths
derived
from
free-radical
cross-
linking
(co)polymerization
processes
[6–9]
as
well
as
monoliths
based
on
(derivatized)
silica-based
materials
[10,11].
Though
some
apparent
conceptual
similarities
in
both
types
of
monoliths
(polymer-
and
silica-based)
can
be
found,
the
mech-
anism
of
interaction,
in
particular
that
of
small
sized
solutes,
differs
profoundly
[12,13].
While
partition
processes
associated
to
the
bonded
layer
positioned
at
the
pore
surface
(or
inside
smaller-sized
pores)
in
silica-based
materials
are
increasingly
understood
in
detail
[14–16],
associated
processes
are
far
more
dominant
and
more
difficult
to
comprehend
in
polymer-based
adsorbents
[12,17–19].
Associated
to
permanently
porous
struc-
tures
of
heterogeneously
cross-linked
polymer
in
the
dry
state,
the
polymer
constituting
the
backbone
of
polymer
monoliths
shows
certain
dynamics
when
operated
in
liquid
chromato-
graphic
applications.
In
contact
with
the
mobile
phases
utilized,
the
heterogeneously
cross-linked
polymer
suffers
from
(solvent-
selective)
nanoscale
solvation
and
swelling
[20,21].
The
resultant
solvent-associated
free
volume
of
polymer
allows
permeation
of
appropriately-sized
solutes
to
a
varying
degree
through
(macro)pore
fluid–gel
interfaces.
This
process
is
primarily
driven
by
partition,
adsorption,
and
associated
diffusion
[22].
Nanoscale
solvation
and
swelling
becomes
affected
by
the
nanoscale
back-
bone
polymer
structure,
nanoscale
distribution
of
cross-link
density,
and
distribution
of
chemistry
within
structural
elements
of
the
monoliths.
Therefore
typical
also,
is
the
variation
of
the
performance
of
polymer
monoliths
with
mobile
phase
compo-
sition
both
for
styrene–divinylbenzene
and
methacrylate-based
chemistries
in
reversed-phase
liquid
chromatographic
applica-
tions
[20,23,24],
as
well
as
in
hydrophilic
interaction
liquid
chromatography
[25].
In
the
past,
fundamental
aspects
of
heterogeneities
found
in
porous
polymer
monoliths
were
discussed
of
having
their
origin
in
a
compositional
drift
of
the
polymerization
precursor
mixtures
and
limited
control
of
free-radical
cross-linking
(co)polymerization
in
material
formation
[23,24].
It
results
in
a
distribution
of
cross-
link
density
as
well
as
associated
chemistry
within
the
monoliths’
internal
backbone
structure
on
a
submicrometer-scale
[21].
We
believe
that
addressing
the
early
events
of
nanoscale
materi-
als’
formation
need
to
be
addressed
as
a
major
key
element
for
the
creation
of
chromatographically
efficient
stationary
phases
[22].
Undoubtedly,
retention
processes
and
resulting
selectiv-
ity
for
separations
have
their
origin
in
a
partition-controlled
process
and
the
materials’
complex
nanoscale
physicochemical
structure.
In
the
search
for
alternate
approaches
to
create
monolithic
materials
offered
by
the
suite
of
techniques
of
polymer
chemistry,
while
retaining
the
technological
simplicity
of
polymer
monolith
preparation,
the
radical-mediated
reactions
of
thiols
with
vinyl
containing
monomers
seems
readily
straightforward.
Such
reac-
tion
may
belong
to
the
family
of
“click”
reactions,
but
only
if
they
follow
a
step-growth
pathway
[26],
an
implementation
most
often
not
realized
in
more
applied
studies
[27–29].
Not
long
ago
we
utilized
polyhedral
oligomeric
vinylsilsesquiox-
anes
(vinylPOSS)
(Fig.
1a)
and
a
diverse
choice
of
thiol
linkers,
two
of
which
selected
in
the
present
study
(Fig.
1b)
[30].
In
situ
mon-
itoring
of
the
reaction
confirmed
an
exclusive
radical-mediated
step-growth
reaction
of
such
precursors.
Such
prerequisite
is
highly
desired
for
the
creation
of
materials
of
a
near
ideal
cross-linked
nanostructure
that
as
well
may
open
new
avenues
for
tailoring
the
microscale
physical
structure
of
porous
adsorbents.
Proper
choice
of
porogenic
diluents
in
the
single-step
molding
process
allowed
for
a
variation
of
microstructural
pore
constitution
and
capability
for
operation
in
liquid
chromatography.
With
the
demonstra-
tion
of
the
creation
of
rigid
and
also
soft
materials
via
such
thiol-ene
“click”
pathway
at
hand
[30,31],
other
processes
related
to
the
preparation
of
hybrid
monolithic
columns
have
been
reported
[32–36].
The
hypothesis
at
the
foundation
of
the
present
report
is
in
experimentally
exploring
rational
design
criteria
for
the
creation
of
hybrid
monolithic
columns
with
near-ideal
nanoscale
networks
based
on
a
radical
process
(Fig.
1c).
This
comes
along
with
suiting
a
straightforward
technological
accessibility
of
macroporous
mono-
lithic
columns
applicable
in
liquid
chromatography.
The
foundation
of
this
experimental
study
can
therefore
be
related
to
our
recent
progress
in
related
materials’
design
approaches
[30,31,37].
In
the
present
study,
we
demonstrate
a
radical-mediated
step-growth
network
creation
which
is
fundamentally
different
from
that
of
typically
utilized
free-radical
chain-growth
and
cross-linking
pro-
cesses
and
marks
the
fundamental
difference
to
materials
created
by
such
processes.
In
addition
to
variable
linker
identities
inter-
connecting
POSS
network
knot
construction
elements
(Fig.
1b),
we
systematically
varied
the
materials
derivation
parameters
such
as
phase
ratio
(controlled
by
monomer
content
of
the
polymerization
mixture),
variation
in
porogenic
solvent
composition
(determining
polymerization-induced
phase
separation
and
microscale
modula-
tion
of
pore
structure)
and
stoichiometry
of
reactants
(determining
cross-link
density
and
nanoscale
network
ideality)
of
resultant
monolithic
entities.
The
resultant
impact
of
all
these
parameters
on
the
hybrid
monoliths’
chromatographic
performance
is
then
detailed
for
the
first
time.
Fundamentally,
this
comes
along
with
limits
and
rationales
for
further
opportunities
in
the
area
of
sepa-
ration
science.
2.
Experimental
2.1.
Chemicals
and
materials
Polyhedral
oligomeric
vinylsilsesquioxane
(vinylPOSS)
cage
mixture
(CH2CHSiO3/2)n(n
=
8,
10,
12)
with
a
nominal
molecular
weight
of
633–950
g/mol,
was
purchased
from
Hybrid
Plastics,
Inc.
(Hattiesburg,
USA).
Azobisisobutyronitrile
(AIBN)
was
purchased
from
Acros
Organics
(Geel,
Belgium).
Concentrated
acetic
acid
(99–100%),
1
M
sodium
hydroxide,
and
1
M
hydrochloric
acid
were
acquired
from
J.T.
Baker
(Deventer,
Holland).
Tetrahydrofuran,
ethanol,
and
acetone
were
purchased
from
VWR
(Fontenay-Sous-
Rois,
France).
All
other
chemicals
were
acquired
from
Sigma
Aldrich
(Vienna,
Austria)
and
used
as
received.
Water
was
purified
on
a
Milli-Q
Reference
water
purification
system
from
Millipore
(Vienna,
Austria).
Teflon
coated
fused-silica
capillaries
of
100
␮m
I.D.
were
pur-
chased
from
Optronis
(Kehl,
Germany)
and
were
used
as
a
mold
for
in
situ,
thermally
initiated
monolith
preparation.
2.2.
Preparation
of
hybrid
monolithic
materials
in
fused-silica
capillaries
and
bulk
A
priori,
the
inner
wall
of
the
100
␮m
I.D.
fused
silica
capillar-
ies
was
provided
with
pendant
methacrylate
functionality
used
to
anchor
the
in
situ
prepared
monolith
to
the
confine
by
an
adapted
procedure
described
previously
[38].
In
detail,
the
inner
surface
of
the
capillaries
was
activated
by
successively
rinsing
it
with
acetone,
water,
0.2
M
sodium
hydroxide
for
1
h,
water,
0.2
M
hydrochlo-
ric
acid
for
1
h,
water,
and
ethanol
by
means
of
a
syringe
pump.
Then,
a
solution
of
50%
of
3-(trimethoxysilyl)propyl
methacrylate
in
ethanol
(v/v)
(adjusted
to
an
apparent
pH
of
5
with
acetic
acid)
was
pumped
through
the
capillary
for
2
h.
Subsequently,
the
capil-
lary
was
flushed
with
ethanol
and
acetone
followed
by
drying
under
a
stream
of
nitrogen.
For
hybrid
column
preparation,
the
capillaries
were
filled
with
the
polymerization
mixtures,
sealed
with
rubber
stoppers,
and

114
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
Fig.
1.
Schematic
representation
of
monomers
and
nanoscale
networks
created
therefrom
through
a
radical-mediated
step-growth
cross-linking
reaction.
(a)
vinylPOSS
monomer,
(b)
two
example
thiol
linkers
with
the
pentaerythritol
tetra(3-mercaptopropionate)
(PETMP)
and
the
2,2-(ethylenedioxy)diethanethiol
(EDDT),
and
(c)
idealized
step-growth
network
formed
through
total
and
equivalent
consumption
of
thiol
and
vinyl
groups
from
equivalent
stoichiometry
(left
panel),
step-growth
network
at
incomplete
though
equivalent
conversion
of
vinyl
and
thiol
groups
(middle
panel),
and
step-growth
network
created
at
an
excess
of
thiol
functional
groups
(right
panel).
placed
in
a
water
bath
tempered
at
60 ◦C
for
16
h.
The
remaining
polymerization
solution
was
filled
in
standard
4
mL
glass
vials
and
allowed
to
polymerize
under
equal
conditions
as
the
capillary
sam-
ples.
For
monolith
preparation,
the
vinylPOSS
cage
mixture
was
first
dissolved
in
tetrahydrofuran
followed
by
addition
of
the
respec-
tive
thiol
and
porogenic
co-solvent
1-dodecanol.
The
homogeneous
liquid-phase
polymerization
mixtures
additionally
contained
AIBN
(1
wt%
with
respect
to
the
thiol).
Further
details
on
the
exact
poly-
merization
mixture
compositions
are
listed
in
Table
1.
After
monolith
formation,
both
ends
of
the
capillary
were
cut
for
attaining
a
final
column
length
close
to
20
cm.
Subsequently,
the
monolithic
columns
were
flushed
with
250
␮L
tetrahydrofuran
and
250
␮L
acetonitrile
in
sequence
at
a
flow
rate
of
1
␮L/min.
Then,
the
columns
were
installed
in
the
nanoLC
instrument
and
equilibrated
with
the
desirable
mobile
phase
composed
of
varying
amounts
of
water
and
acetonitrile.
Bulk
polymerized
materials
were
soxhlet-
extracted
overnight
with
tetrahydrofuran,
followed
by
drying
and
further
analysis.
2.3.
Apparatus,
determination
of
gel
fraction,
and
chromatographic
measurements
Appropriate
samples
of
bulk
monoliths
underwent
spectro-
scopic
investigations.
Attenuated
total
reflection-Fourier
transform
infrared
spectroscopy
(ATR-FTIR)
was
performed
on
a
PerkinElmer
Spectrum
100
spectrometer
(PerkinElmer
Vertriebs
GmbH,
Vienna,
Austria).
Raman
spectra
were
acquired
using
a
portable
i-
Raman
Plus
B&WTek
spectrometer
(Polytec,
Waldbronn,
Germany)
equipped
with
a
785
nm
laser.
A
laser
intensity
of
100%
was
used
in
all
experiments.
Scanning
electron
microscopy
(SEM)
of
the
capillary
columns’
cross-section
was
performed
on
a
Crossbeam
1540
XB
electron
microscope
(Carl
Zeiss
SMT
AG,
Oberkochen,
Germany).
Thermogravimetric
analysis
of
bulk
monoliths
was
carried
out
on
a
TA
Q5000
instrument
(Waters
GmbH,
Eschborn,
Germany)
under
air
atmosphere
at
a
heating
ramp
of
10 ◦C/min
from
T
=
50 ◦C
to
800 ◦C.
Nitrogen
sorption
measurements
were
carried
out
with
a
Micromeritics
TriStar
II
Surface
Area
and
Porosity
Instrument
(SY-
LAB
Geräte
GmbH,
Neu-Pukersdorf,
Austria).
The
gel
fraction
of
monolithic
materials
was
determined
from
bulk
polymerizations,
in
which
the
weight
of
the
dried
material
after
soxhlet
extraction
was
divided
by
the
weight
of
the
precursors
used
for
material
formation
according
to
the
following
equation:
Gel
fraction
=Weightmaterial,dry
Weightprecursors
×
100%
(1)
Chromatographic
experiments
with
capillary
monoliths
were
performed
on
a
modulated
Dionex
Ultimate
3000
nanoLC
system
(Dionex
GmbH,
Vienna,
Austria),
incorporating
a
flow
splitter,
a
flow
sensor,
and
a
flow
control
valve.
Injections
were
enabled
by
a
Vici
Valco
Cheminert
4
nL
internal
sample
loop
injection
valve
(Bartelt
GmbH,
Vienna,
Austria)
switched
in
line
with
the
flow
path.
The
UV-detector
cell
volume
was
3
nL
which,
together
with
the
transfer
capillary
connected
to
the
column
outlet,
resulted
in
a
min-
imal
post-column
volume
of
104
nL.
UV-detection
was
carried
out
at
210
nm.
For
isocratic
separations
at
variable
flow
rates
and
mobile
phase
compositions,
standard
mixtures
of
uracil,
benzyl
alcohol,
ben-
zene,
and
five
alkyl
benzenes
were
dissolved
in
the
running
mobile

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
115
Table
1
Polymerization
mixture
compositions
and
dry-state
properties
of
all
bulk
monolithic
materials
prepared
from
identical
polymerization
formulations
as
those
used
for
capillary
monolith
preparation.
Monolith
ThiolaMonomer
amountb(%,
w/w)
Vinyl
to
ThiolcSolventd(%,
w/w)
Gel
fractione
(%)
Surface
areaf
(m2/g)
Ceramic
yield
expectedg(%)
Ceramic
yield
obtainedh(%)
1
PETMP
20
1:1
50/50
95.2
3.6
29.9
28.9
1a
PETMP
30
1:1
50/50
96.6
10.2
29.9
28.3
1b
PETMP
30
1:1.5
50/50
91.9
2.0
22.9
24.1
1c
PETMP
30
1:1
45/55
96.6
2.2
29.9
29.1
2
EDDT
20
1:1 40/60
89.7 1.1 35.8
35.0
2a
EDDT
25
1:1
35/65
91.5
2.2
35.3
36.5
aLinker
utilized
according
to
Fig.
1b,
PETMP
(pentaerythritol
tetra(3-mercaptopropionate))
and
EDDT
(2,2-(ethylenedioxy)diethanethiol).
bTotal
amount
of
monomers
in
the
polymerization
mixture.
cMolar
ratio
of
vinyl-to-thiol
functional
groups
present
in
the
polymerization
mixture.
dPorogenic
solvent
composition
of
tetrahydrofuran/dodecanol
utilized.
eCalculated
according
to
Eq.
(1).
Typically,
the
gel
fraction
did
not
vary
more
than
±0.4%.
fDry-state
Brunauer–Emmett–Teller
(BET)
surface
areas
determined
from
nitrogen
sorption
experiments.
gExpected
estimate
based
on
the
theoretical
composition
of
the
step-growth
network
assuming
that
all
SiO3/2 content
theoretically
present
in
these
networks
is
oxidized
to
SiO2at
the
end
of
a
thermogravimetric
run.
hEstimates
from
thermogravimetric
runs
in
air
to
T
=
800 ◦C
(Fig.
S2).
Residues
left
were
found
amorphous
silica
by
FTIR
analysis.
phases.
Typically,
the
chromatographic
performance
of
columns
estimated
with
these
solutes
and
across
studied
velocities
did
not
vary
more
than
±2.5
␮m.
The
superficial
velocity
based
hydrodynamic
permeability
and
total
porosity
was
determined
as
reported
recently
[18].
3.
Results
3.1.
Overview
of
monolith
preparation
In
this
study,
vinylPOSS
(Fig.
1a)
was
selected
as
a
key
building
block
for
the
construction
of
porous
polymer
monolithic
columns
to
use
in
nano-LC
by
radical
mediated
step-growth
thiol-ene
chem-
istry.
Two
different
types
of
monoliths
were
generated
by
select-
ing
two
chemically
distinct
thiol-containing
monomers,
i.e.
the
pentaerythritol
tetra(3-mercaptopropionate)
(PETMP)
and
the
2,2-(ethylenedioxy)diethanethiol
(EDDT)
(Fig.
1b).
In
the
PETMP,
the
thiol
groups
are
preceded
by
ester
groups
(electron
withdraw-
ing),
and
in
the
EDDT
the
thiol
groups
are
preceded
by
ether
groups
(electron
donating).
In
other
words,
thiol
groups
in
the
PETMP
possess
a
more
electrophilic
nature
than
thiol
groups
in
the
EDDT
[39].
Based
on
their
chemical
structure,
the
hydrophobic–hydrophilic
character
of
these
monomers
is
expected
similar.
This
allowed
using
similar
porogenic
solvents
and
their
compositions.
Such
approach
is
important
since
the
porogenic
solvents
allow
for
mod-
ulating
polymerization-induced
phase
separation
processes
and
therefore
creation
of
desirable
flow-through
pore
structures
within
the
monoliths.
It
also
alleviates
comparison
of
the
nanoscale
back-
bone
structure
of
the
monoliths
and
classification
of
morphology
of
flow-through
pores
on
the
materials’
studied
chromatographic
performance
in
a
straightforward
manner.
Careful
choice
of
porogenic
solvent
compositions
enabled
cre-
ation
of
a
porous
flow-through
pore
structure
that
allowed
for
liquid
flow
at
decent
backpressures
and
a
robust
chromatographic
performance.
The
exactly
utilized
polymerization
mixture
compo-
sitions
including
amounts
of
tetrahydrofuran
and
1-dodecanol
as
porogenic
diluents
can
be
found
in
Table
1.
Chemically,
we
worked
at
an
equivalent
stoichiometry
of
vinyl-
to-thiol
functional
groups
in
the
preparation
of
hybrid
monoliths
at
different
ratio
of
monomers
to
porogenic
solvents,
i.e.
creating
different
phase
ratios
(Monoliths
1
and
1a,
Monoliths
2
and
2a,
Table
1).
Additionally,
we
also
worked
at
an
imbalance
of
stoichi-
ometric
ratios
of
vinyl-to-thiol
functional
groups
to
create
hybrid
monoliths
with
the
PETMP
linker
(Monolith
1b)
in
order
to
assess
the
impact
of
a
varied
degree
of
cross-linking.
For
the
study
of
a
varied
flow-through
pore
structure,
we
as
well
varied
the
macro-
pore
size
and
associated
flow-through
pore
structure
of
Monolith
1a
through
adjustment
of
the
porogenic
solvent
composition
under
otherwise
same
preparatory
conditions
resulting
in
Monolith
1c.
3.2.
Dry-state
structural
and
compositional
assessment
of
the
materials
and
surface
area
from
nitrogen
sorption
measurements
To
obtain
rational
insight
into
the
chemistry
and
structure
of
the
materials
as
reflected
by
the
dry
state,
we
performed
analysis
from
prepared
samples
in
bulk.
These
samples
are
based
on
the
same
polymerization
mixtures
utilized
for
creation
of
monolithic
structures
in
the
capillaries
(Table
1).
Though
the
dimensions
of
the
confine
vary
largely,
polymer
preparation
appears
most
closely
mimicked
to
that
of
the
capillaries.
ATR-FTIR
spectra
of
Monoliths
1
and
1a
prepared
at
equivalent
stoichiometry
of
vinyl-to-thiol
groups
revealed
no
signals
associ-
ated
to
the
C
C
stretching
(1600
cm−1),
nor
to
the
S
H
stretching
(2570
cm−1)
(Fig.
2a).
On
the
other
hand,
Monoliths
2
and
2a
evidenced
a
very
small
stretching
at
1600
cm−1(Fig.
2b).
This
pic-
ture
is
confirmed
with
the
Raman
spectra
(Figs.
S1a
and
b).
Here,
small
stretching
signals
for
vinyl
and
thiol
groups
could
be
found
in
spectra
of
both
monoliths,
in
particular
for
Monoliths
2
and
2a.
The
comparably
lower
peak
intensities
of
thiol
and
vinyl
groups
found
in
Monoliths
1
and
1a
in
comparison
to
Monoliths
2
and
2a
indicate
a
higher
extent
of
the
actual
cross-linking
reaction
for
the
former.
In
light
of
previous
investigations
we
have
found
that
the
here
presented
principal
reaction
follows
a
step-growth
pathway
[30].
This
becomes
supported
by
Fig.
S1a,
where
at
a
stoichiometric
bal-
ance
of
vinyl-to-thiol
groups
in
the
created
monoliths,
small
signals
for
both
functional
groups
coexist.
In
the
case
of
an
excess
of
thiol
(Monolith
1b)
the
signal
for
the
vinyl
stretching
entirely
vanishes
together
with
a
more
pronounced
signal
for
the
thiol
(2570
cm−1).
Change
in
porogenic
solvents
at
equimolarity
of
functional
groups
with
the
example
of
Monolith
1c
indicates
very
similar
spectral
fea-
tures
than
that
of
Monolith
1a.
The
otherwise
equivalent
monomer
concentration
and
stoichiometry
in
the
preparation
of
Monoliths
1a
and
1c
is
a
straightforward
explanation.
To
gain
further
insight
into
the
above-described
observations,
we
determined
the
gel
fractions
for
the
created
materials
(Eq.
(1)).
This
is
a
useful
estimate
to
judge
on
the
efficiency
of
transfor-
mation
of
a
known
amount
of
monomeric
precursors
to
that
of
their
presence
in
the
formed
material.
Gel
fractions
found
were
95.2
and
96.6%
for
Monoliths
1
and
1a
and
89.7
and
91.5%
for
Monoliths
2
and
2a
(Table
1).
These
observations
corroborate
obser-
vations
by
spectroscopy
that
additionally
indicated
a
more
efficient

116
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
Fig.
2.
ATR-FTIR
spectra
of
(a)
PETMP
(blue
line),
vinylPOSS
(black
line),
Monolith
1
(orange
line),
and
Monolith
1a
(red
line);
(b)
EDDT
(blue
line),
vinylPOSS
(black
line),
Monolith
2
(orange
line),
and
Monolith
2a
(red
line).
(For
interpretation
of
the
references
to
color
in
this
figure
legend,
the
reader
is
referred
to
the
web
version
of
this
article.)
cross-linking
for
monoliths
created
with
PETMP
as
linker,
i.e.
almost
non-detectable
vinyl
and
thiol
groups
left
within
the
materials
(Fig.
2
and
Fig.
S1).
This
as
well
supports
that
the
PETMP
is
more
prone
to
form
cross-links
between
individual
POSS
cages
and
that
its
reactivity
is
superior
to
that
of
EDDT.
This
situation
is
not
sur-
prising
due
to
the
reactivity
of
ester
type
of
thiols
[30,31].
Based
on
this
discussion,
it
is
therefore
also
safe
to
assume
that
the
hybrid
networks
created
with
EDDT
as
the
linking
unit
show
larger
deviation
from
an
idealized
step-growth
network
(Fig.
1c,
middle
panel)
than
hybrid
networks
created
with
the
PETMP
(Fig.
1c,
left
panel).
Though
small,
an
increase
in
gel
frac-
tion
from
Monoliths
1
to
1a
and
from
Monoliths
2
to
2a
is
indicated
(Table
1).
This
suggests
an
overall
increased
functional
group
con-
version
for
monoliths
created
with
an
increased
phase
ratio,
i.e.
an
increased
overall
density
of
porous
material
created.
Monolith
1b
prepared
with
an
excess
of
thiol
showed
a
rela-
tively
lower
gel
fraction
when
compared
to
Monolith
1a
(Table
1).
This
agrees
with
our
previous
observations
with
related
materi-
als,
where
obtained
gel
fractions
were
also
highest
at
equimolar
amounts
of
vinyl
and
thiol
groups,
or
excess
of
vinyl
groups
located
on
the
POSS
[31].
Therefore
also,
when
using
an
excess
of
thiol,
the
nanostructure
of
Monolith
1b
is
more
closely
resembled
by
Fig.
1c
(right
panel).
Monolith
1c
exhibited
a
similar
gel
fraction
as
Monolith
1a.
This
is
expected
when
a
similar
amount
of
monomers
and
equivalent
stoichiometry
is
present
in
the
polymerization
mixture.
However,
the
variable
porogenic
solvent
composition
allows
for
tailoring
large
pores
protruding
the
material
(vide
infra).
TGA
analysis
in
air
revealed
that
both
types
of
monoliths
expe-
rience
degradation
above
T
=
300 ◦C
(Fig.
S2).
All
monoliths’
silica
residues
obtained
at
the
end
of
this
analysis
showed
very
good
agreement
to
theoretically
expected
values
(Table
1).
This
corrob-
orates,
alongside
spectral
data
(Fig.
2
and
Fig.
S1)
and
gel
fractions
around
≥90%,
the
attainment
of
step-growth
networks
whose
com-
position
is
determined
by
the
chosen
monomers
and
stoichiometric
ratios
used
for
material
formation.
Not
surprisingly,
nitrogen
sorption
analysis
of
the
densely
cross-linked
macroporous
materials
revealed
calculated
dry-state
Brunauer–Emmett–Teller
(BET)
surface
areas
not
exceeding
those
typically
observed
for
polymer
monolithic
columns,
consistent
with
earlier
reported
results
[30].
This
as
well
indicates
an
almost
complete
absence
of
permanent
micro-/mesoporosity.
However,
it
is
noticeable
that
monoliths
created
with
PETMP
always
showed
superior,
though
still
very
small,
dry
state
surfaces
areas
com-
pared
to
the
monoliths
created
with
EDDT.
Though
within
a
small
range,
specific
surface
areas
were
enhanced
at
higher
phase
ratios
of
the
created
monoliths
clearly
indicated
for
Monoliths
1
and
1a
(Table
1).
3.3.
Capillary
scale
morphology
The
constitution
of
flow-through
pore
space
is
very
important
for
liquid
chromatographic
applications
since,
first
of
all,
it
deter-
mines
the
efficiency
with
which
solutes
can
be
transported
through
the
porous
structures
without
significant
adsorption
and
partition
processes
[12].
The
dry-state
structure
of
the
hybrid
monoliths
was
investigated
by
SEM
(Fig.
3).
SEM
indicated
that
all
mono-
liths
studied
were
well-anchored
to
the
inner
wall
of
the
fused
silica
capillaries
and
showed
a
much
denser
and
more
homoge-
neous
overall
appearance
for
both
column
types
at
increased
phase
ratio,
but
maintained
stoichiometry
(Monoliths
1,
1a
and
Mono-
liths
2,
2a).
This
is
not
surprising
due
to
the
larger
amount
of
monomers
and
initiator
leading
to
a
larger
amount
of
initiation
sites
and
a
more
effective
filling
of
capillary
cross-sectional
space.
The
columns’
morphology
shows
certain
variations.
However,
the
majority
of
them
consist
of
a
skeleton
of
aggregated
microstructural
elements.
The
size
of
these
structural
features
appears
uniform
along
cross-sections,
in
particular
for
Monoliths
1a
and
2a.
The
morphology
of
Monoliths
1b,
1c,
and
2a
is
more
pro-
nouncedly
globular.
In
particular,
monoliths
that
contrast
to
the
poorly
organized
(cauliflower-like)
globular
topology
typically
observed
for
highly
permeable
conventional
polymer
monoliths
are
very
interesting
(vide
infra)
[18,23,40].
The
more
continuous-like
skeletal
features,
in
particular
found
for
Monolith
1a,
are
reminis-
cent
of
silica-based
[41,42],
and
organic-silica
hybrid
monolithic
columns
[43].
The
hybrid
Monolith
2
based
on
the
EDDT
linker
exhibited
a
visually
finer
skeleton
than
the
corresponding
monolith
prepared
with
the
PETMP
linker
(Monolith
1),
a
result
of
reaction
kinetics
and
specifics
of
phase
separation.
Monoliths
1b,
prepared
at
stoi-
chiometric
disparity
of
reactive
functional
groups
and
Monolith
1c,
prepared
at
increased
amount
of
dodecanol
replacing
tetrahy-
drofuran
in
the
porogenic
solvent
mixture,
respectively,
exhibited
clearly
larger
structural
feature
sizes
(Fig.
3c
and
d)
than
the
corresponding
Monolith
1a
(Fig.
3a,
lower
images).
With
these
qual-
itative
observations
at
hand,
we
move
to
the
performance
under
non-retained
elution
conditions.

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
117
Fig.
3.
SEM
of
capillary
cross-sections
and
respective
zoom
area
of
interest
for
(a)
Monolith
1
(upper
images)
and
Monolith
1a
(lower
images);
(b)
Monolith
2
(upper
images)
and
Monolith
2a
(lower
images);
(c)
Monolith
1b;
(d)
Monolith
1c.
3.4.
Performance
under
non-retained
elution
conditions,
chromatographic
retention,
and
sensitivity
of
performance
on
retention
3.4.1.
Non-retained
elution
conditions
and
impact
of
material
density
on
performance
With
the
comprehensive
data
of
the
previous
sections
at
hand,
we
studied
the
elution
performance
with
non-retained
uracil
on
all
columns
(Fig.
4)
at
a
mobile
phase
composition
of
50/50
ace-
tonitrile/water
(%,
v/v).
It
is
seen
for
both
cases
of
variation
of
phase
ratio
that
monolithic
columns
prepared
with
the
EDDT
(Monoliths
2
and
2a)
showed
a
better
performance
than
their
PETMP
based
counterparts
(Monoliths
1
and
1a)
(Fig.
4a).
How-
ever,
in
both
cases
monoliths
at
increased
phase
ratio
show
better
performance.
Under
non-retained
conditions,
Monolith
1b
performs
very
close
to
Monolith
1a,
while
a
slightly
larger
deterioration
of
effi-
ciency
at
increased
flow
velocities
can
be
observed
for
Monolith
1c
(Fig.
4b).
The
overall
relative
insensitivity
of
performance
for
the
non-retained
tracer
uracil,
but
obvious
difference
in
micro-
structural
constitution
(as
indicated
in
Fig.
3),
once
more
confirms
the
limited
insight
a
non-retained
tracer
can
provide
for
the
mate-
rials’
performance
in
a
practical
scenario
[23].
The
here
reported
efficiencies
still
do
not
reach
those
of
state-
of-the-art
capillary-scale
silica-based
monoliths
[44].
However,
performance
is
very
much
enhanced
compared
to
that
of
ana-
lytical
format
porous
polymer
monoliths
[13,45–49]
and
also
capillary-scale
polymer
monolithic
columns
based
on
free-radical
cross-linking
(co)polymerization
[23,24,50].
Interestingly,
even
under
non-retained
conditions
we
see
a
“pseudo”
C-Term
for
uracil,
indicating
some
impact
of
microstructural
heterogeneity
[19,51,52].
This
impact
is
significantly
reduced
at
increased
phase
ratio
(Fig.
4a)
and
a
corresponding
superficially
more
homoge-
neous
capillary-scale
cross-sectional
porous
structure
(Fig.
3a
and
b).
Therefore,
the
reported
data
are
a
useful
reference
for
our
studies
on
retention
and
efficiency
processes
since
it
is
known
that
the
real-
istic
performance
for
related
types
of
materials
ostensibly
requires
the
use
of
retained
probe
solutes
[13].
In
the
following,
we
report
retention
properties
of
the
materials
first
and
then
sensitivity
of
determined
retention,
mobile
phase
composition,
and
flow
velocity
on
the
plate
height
of
small
retained
solutes.
3.4.2.
Chromatographic
retention,
permeability,
and
porosity
Despite
their
decently
small
dry-state
surface
areas
(Table
1),
all
monoliths
reported
in
the
present
study
provided
substantial
retention
and
selectivity
in
the
elution
of
benzyl
alcohol,
ben-
zene,
and
alkyl
benzenes,
i.e.
behavior
qualifying
these
materials
as
reversed-phase
type
of
stationary
phases.
Fig.
5
shows
reten-
tion
of
the
here
prepared
monoliths
(Table
1)
at
a
mobile
phase
composition
of
50/50
acetonitrile/water
(%,
v/v).
In
cases
of
an
increase
in
phase
ratio,
we
observe
a
substan-
tial
increase
in
retention
for
benzene,
and
the
five
alkyl
benzenes
(Fig.
5a).
At
equivalent
phase
ratio
we
find
the
retention
of
Monolith
1
larger
than
that
of
Monolith
2,
while
an
increased
phase
ratio
in
both
cases
maintains
overall
methylene
selectivity.
This
selectivity
allows
for
baseline
resolution
of
solutes
differing
in
just
one
methy-
lene
unit
with
room
for
additional
peak
capacity
(Fig.
S3)
even
at
velocities
higher
than
the
optimum
of
the
plate
height
curve
(vide
infra).
At
increasing
phase
ratio,
e.g.
when
moving
from
Monolith
1
to
Monolith
1a,
we
observed
an
average
solute
retention
increase
of
about
60%
at
a
similar
selectivity
(Fig.
5a,
empty
vs.
filled
squares).
This
increased
phase
ratio
is
accompanied
with
a
reduced
perme-
ability
and
porosity
from
Monolith
1
(kp,f =
9
×
10−14 m2,
εt=
0.93)
to
Monolith
1a
(kp,f =
1.2
×
10−14 m2,
εt=
0.84).
From
Monolith
2
to
Monolith
2a
the
average
solute
retention
increase
was
around
50%,
again
at
apparently
similar
selectivity
(Fig.
5a,
empty
vs.
filled
circles).
Here,
the
increased
phase
ratio
and
consequently
decreased
porosity
do
not
appear
to
influence
permeability
much,
a
reason
found
in
the
slightly
different
poro-
genic
solvent
composition
employed
(Table
1).
The
permeability
and
porosity
values
for
Monolith
2
(kp,f =
11
×
10−14 m2,
εt=
0.96)
and
for
Monolith
2a
(kp,f =
14
×
10−14 m2,
εt=
0.89)
support
this
statement.

118
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
Fig.
4.
Performance
under
non-retained
elution
conditions
at
a
mobile
phase
com-
position
of
50/50
acetonitrile/water
(%,
v/v)
probed
with
uracil
for
(a)
Monolith
1
(filled
squares),
Monolith
1a
(empty
squares),
Monolith
2
(filled
circles),
Monolith
2a
(empty
circles);
(b)
Monolith
1a
(empty
squares),
Monolith
1b
(filled
triangles),
Monolith
1c
(empty
triangles).
The
increase
in
retention
with
increased
phase
ratio
is
in
qualita-
tive
agreement
to
what
is
typically
observed
for
polymer
monolith,
i.e.
scaling
of
retention
with
the
phase
ratio
due
to
an
increase
of
material
in
the
confine
that
can
provide
retention.
This
is
almost
quantitatively
observable
when
PETMP
is
used
as
the
linker
(Mono-
liths
1
and
1a).
Interestingly,
an
imbalance
in
stoichiometry
at
the
same
overall
wt%
of
monomers
in
the
polymerization
mixtures
for
the
mono-
liths
prepared
with
PETMP
as
linker
(Monolith
1b
vs.
1a)
led
to
a
significant
reduction
in
retention
(Fig.
5b).
This
is
because
at
an
excess
of
thiol,
a
lower
amount
of
hydrophobic
POSS
is
incorporated
in
the
network
structure.
As
well
we
note
a
significantly
reduced
gel
fraction
(Table
1),
indicating
that
not
all
thiol
monomers
are
incorporated
in
the
network.
The
resultant
reduced
degree
of
cross-linking
apparently
leads
to
larger
structural
features
and
con-
sequently
to
increased
permeability
and
slightly
increased
porosity
from
Monolith
1a
(kp,f =
1.2
×
10−14 m2,
εt=
0.84)
to
Monolith
1b
(kp,f =
11
×
10−14 m2,
εt=
0.87).
Replacing
5
wt%
of
tetrahydrofuran
by
the
poorer
solvent
1-
dodecanol
at
the
same
overall
wt%
of
monomers
and
stoichiometry
used
to
prepare
Monolith
1a,
substantially
increased
permeabil-
ity
of
Monolith
1c
(kp,f =
16
×
10−14 m2)
at
comparable
porosity
(εt=
0.86).
This
is
due
to
an
earlier
phase
separation
and
becomes
supported
by
the
SEM
images
that
indicate
a
larger
flow-through
Fig.
5.
Retention
factor
against
number
of
alkyl
carbon
atoms
of
the
homologous
series
of
benzene,
and
five
alkylbenzenes
at
a
mobile
phase
composition
of
50/50
acetonitrile/water
(%,
v/v).
(a)
Monolith
1
(filled
squares),
Monolith
1a
(empty
squares),
Monolith
2
(filled
circles),
Monolith
2a
(empty
circles);
(b)
Monolith
1a
(empty
squares),
Monolith1b
(filled
triangles);
(c)
Monolith
1a
(empty
squares),
Monolith
1c
(empty
triangles).
pore
space
and
associated
feature
sizes
of
the
monolith’s
back-
bone
(Fig.
3d
in
contrast
to
Fig.
3a,
lower
images).
Owing
to
the
same
phase
ratio,
gel
fraction,
and
composition
obtained
from
the
same
stoichiometry
of
precursors
utilized
for
preparation
(Table
1,
Fig.
S1),
retention
is
only
moderately
influenced
(Fig.
5c).
The

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
119
Fig.
6.
(a)
Plate
height
curves
measured
on
Monolith
1
(gray
symbols)
and
Mono-
lith
1a
(black
symbols)
with
uracil
(squares),
benzyl
alcohol
(circles),
benzene
(triangles),
toluene
(diamonds),
ethylbenzene
(half-filled
squares),
propylbenzene
(half-filled
circles),
butylbenzene
(half-filled
triangles),
and
pentylbenzene
(half-
filled
diamonds);
(b)
plate
height
versus
retention
factor
of
the
probe
solutes
at
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s
for
Monolith
1
(gray
filled
circles)
and
Monolith
1a
(black
filled
circles).
The
mobile
phase
composition
was
50/50
acetonitrile/water
(%,
v/v).
substantially
different
morphology
and
flow-through
pore
struc-
ture
of
Monolith
1c
(Fig.
3d)
in
contrast
to
Monolith
1a
(Fig.
3a,
lower
images)
allows
us
for
a
desirable
comparison
of
the
impact
of
morphology
on
performance
for
non-retained
and
in
particular
retained
solutes
(vide
infra).
3.4.3.
Chromatographic
performance
–
impact
of
materials’
density,
thiol
linker,
and
retention
with
varying
ideality
of
nanoscale
hybrid
polymer
networks
Early
in
experiments
we
noted
that
mixtures
of
uracil
as
non-
retained
tracer,
benzyl
alcohol
as
slightly
retained
tracer,
benzene,
and
five
alkyl
benzenes
could
successfully
be
baseline
separated
on
all
monoliths
prepared
in
this
study
under
simple
isocratic
mobile
phase
conditions
(with
example
separations
of
Monoliths
1a
and
2a
shown
in
Fig.
S3).
In
order
to
assess
the
chromatographic
per-
formance
in
the
separation
of
small
molecules
with
the
hybrid
step-growth
polymeric
monoliths
here
prepared,
we
have
utilized
this
solute
mixture
at
a
constant
mobile
phase
composition
of
50/50
acetonitrile/water
(%,
v/v)
for
all
columns.
Figs.
6a
and
7a
show
the
plate
height
curves
for
Monoliths
1
and
1a
and
for
Monoliths
2
and
2a,
respectively.
It
can
be
clearly
observed
that
Monoliths
1a
and
2a
performed
superior
to
their
Fig.
7.
(a)
Plate
height
curves
measured
on
Monolith
2
(gray
symbols)
and
Monolith
2a
(black
symbols)
with
symbol
assignments
to
solutes
as
in
Fig.
6;
(b)
plate
height
versus
retention
factor
at
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s
for
Monolith
2
(gray
filled
circles)
and
Monolith
2a
(black
filled
circles).
Same
mobile
phase
as
in
Fig.
6.
corresponding
lower
phase
ratio
counterparts,
in
particular
in
the
right
hand
branch
of
the
plate
height
curve.
This
already
became
indicated
by
studies
with
the
non-retained
uracil
only
(Fig.
4)
together
with
that
of
SEM
images
(Fig.
3).
Similar
observations
were
also
made
by
Tanaka’s
group
when
preparing
silica
monoliths
at
increased
phase
ratios
[44].
Besides
superior
efficiency,
we
can
also
observe
that
the
varia-
tion
of
efficiency
with
solute
identity
(and
consequently
retention)
is
clearly
different
for
these
columns.
The
retention-dependence
of
efficiency
is
very
much
attenuated
in
particular
for
Monolith
1a
compared
to
Monolith
1
(Fig.
6a)
while
individual
solutes’
plate
height
curves
are
always
more
pronouncedly
distinct
for
Monoliths
2a
and
2
(Fig.
7a).
In
order
to
rationalize
this
observation,
we
had
a
closer
look
on
how
individual
solutes’
efficiency
varies
at
identical
mobile
phase
composition
and
with
two
example
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s
for
Monoliths
1
and
1a
(Fig.
6b)
and
for
Monoliths
2
and
2a
(Fig.
7b).
At
a
flow
velocity
of
0.2
mm/s,
efficiency
appeared
similar
for
probe
solutes
in
any
of
these
mono-
liths.
This
velocity
is
close
to
the
optimum
of
the
plate
height
curves.
However,
at
a
higher
flow
velocity
of
1.7
mm/s,
clearly
entering
a
flow
velocity
range
that
may
significantly
include
resistive
contrib-
utions
to
mass
transfer
[50],
Monoliths
1
and
2
clearly
resemble
solute-dependent
and,
therefore,
retention-dependent
efficiency.

120
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
For
Monolith
1a,
this
retention-dependent
efficiency
is
virtually
absent
as
compared
to
its
lower
phase
ratio
counterpart
Monolith
1
(Fig.
6b).
This
monolith
as
well
has
the
highest
gel
fraction
and
most
idealized
nanoscale
hybrid
polymer
networks
(Fig.
1c,
left
panel).
For
Monolith
2a,
a
significant
variation
of
the
efficiency
with
the
solutes’
identity
can
still
be
observed
(Fig.
7b)
despite
its
increased
phase
ratio.
This
is
in
line
with
a
nanoscale
non-ideality
of
cross-
linked
networks
being
still
persistent
at
higher
phase
ratios
and
lower
gel
fractions
(Table
1)
when
using
the
EDDT
linker
(Fig.
1c,
middle
panel).
Though
the
EDDT
linker
provides
slightly
better
efficiency
under
non-retained
elution
conditions
than
columns
pre-
pared
with
the
PETMP
linker
(Fig.
4a),
the
retention
dependency
of
performance
clearly
makes
this
column
overall
inferior
since
the
EDDT
linker
shows
a
clearly
noticeable
variation
of
performance
among
small
retained
solutes
(Fig.
7b).
Overall,
these
important
observations
demonstrate
a
han-
dle
enabling
for
control
of
key
obstacles
typically
observed
with
polymer
monoliths
created
by
free
radical
cross-linking
(co)polymerization
including
commercial
products.
Such
mono-
liths
are
characterized
by
heterogeneous
nanoscale
cross-linked
polymer
networks
[21],
while
a
tailoring
toward
more
idealized
nanoscale
network
structures
apparently
enhances
performance,
in
particular
for
retained
solutes,
in
the
present
work.
3.4.4.
Chromatographic
performance
–
impact
of
stoichiometry
In
order
to
test
our
initial
hypothesis
and
based
on
above
observations
that
the
tailoring
of
(near)
ideal
nanoscale
network
structures
(Fig.
1c,
left
panel)
is
a
determining
prerequisite
when
designing
advanced
monoliths,
we
have
deliberately
introduced
nanoscale
network
defects
(Fig.
1c,
right
panel).
We
have,
therefore,
taken
the
best
performing
Monolith
1a,
and
adjusted
stoichiom-
etry
of
reactants.
We
chose
a
ratio
of
vinyl-to-thiol
functional
groups
of
1:1.5
(Table
1)
while
keeping
the
same
overall
amount
of
monomers
(%,
w/w)
in
the
polymerization
mixture
at
identical
porogenic
solvent
composition
(Monolith
1b).
When
compared
to
Monolith
1a,
this
disparity
in
stoichiometry
must
result
in
a
reduced
degree
of
cross-linking
density
within
the
step-growth
network.
This
becomes
further
amplified
by
resulting
reduced
gel
fractions
(Table
1)
[31].
This
reduced
cross-link
density
is
clearly
supported
by
the
Raman
spectra
(Fig.
S1a).
On
the
monolith’s
nanoscale,
an
increased
amount
of
dangling
thiol
arms
from
the
PETMP
monomer
must
be
present
and
impart
nanoscale
network
defects.
Such
statis-
tical
introduction
of
network
defects
may
impact
chromatographic
performance
as
well
(Fig.
1c,
right
panel
as
opposed
to
left
panel).
Fig.
8a
shows
how
the
performance
of
Monolith
1b
slightly
deteriorates
in
relation
to
that
of
Monolith
1a
for
non-retained
uracil.
The
efficiency
for
retained
solutes
decreases
more
pro-
nouncedly
on
the
right
hand
side
of
the
plate
height
curves
with
a
clearly
observed
retention-dependent
efficiency
(Fig.
8b).
While
this
change
in
performance-retention
dynamics
is
less
noticeable
at
a
low
linear
chromatographic
flow
velocity
of
0.2
mm/s,
it
is
more
pronounced
at
increased
linear
chromatographic
flow
velocities
of
1.7
mm/s,
i.e.
in
the
regime
that
is
dominated
by
mass
transfer
resistance
contributions.
The
microstructural
morphology
found
for
Monolith
1b
was
also
indicated
to
be
composed
of
larger
structural
features
than
those
constituting
Monolith
1a
(Fig.
3c
in
comparison
to
Fig.
3a,
lower
images).
Larger
structural
elements
are
additionally
known
to
provide
greater
penetration
depth
for
solutes
based
on
par-
tition
and
adsorption
[18].
This
partition
and
adsorption
may
hamper
mass
transfer
efficiency,
seen
in
particular
at
higher
flow
velocities
and
for
the
retained
solutes.
In
addition,
these
struc-
tural
elements
deliberately
possess
an
overall
lower
degree
of
cross-linking
(Fig.
1c,
right
panel).
The
structural
modification
certainly
enhances
the
solvent-associated
polymer
free
volume
accessed
by
solutes
and
contributes
to
some
deterioration
in
mass
Fig.
8.
(a)
Plate
height
curves
measured
on
Monolith
1b
(black
symbols)
in
contrast
to
Monolith
1a
(gray
symbols)
with
symbol
assignments
to
solutes
as
in
Fig.
6;
(b)
plate
height
versus
retention
factor
at
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s
for
Monolith
1b
(black
filled
circles)
in
contrast
to
Monolith
1a
(gray
filled
circles).
Same
mobile
phase
as
in
Fig.
6.
transfer
efficiency,
a
result
supporting
previous
investigations
based
on
purely
polymeric
counterparts
[18,24].
3.4.5.
Chromatographic
performance
–
impact
of
microstructural
feature
size
In
order
to
better
assess
the
interplay
between
microstructural
morphology
with
these
newly
derived
materials,
we
proceed
with
characterization
of
Monolith
1c,
whose
polymerization
mixture
formulation
only
differs
from
that
used
for
Monolith
1a
by
an
incre-
mental
amount
of
5
wt%
of
1-dodecanol
replacing
tetrahydrofuran
in
the
porogenic
solvent
system
(Table
1).
Under
these
conditions,
a
microstructural
morphology
composed
of
relatively
larger
struc-
tural
features
and
larger
macropores
was
obtained
(Fig.
3d).
As
was
already
mentioned
before,
such
morphology
may
provide
resis-
tance
to
mass
transfer,
in
particular
for
retained
solutes.
When
assessing
the
molecular
structure
of
Monolith
1c
from
bulk
material,
we
concluded
on
a
very
similar
polymer
nanostruc-
ture
to
that
of
Monolith
1a,
since
equivalent
gel
fractions,
inorganic
residues
from
thermogravimetric
analysis,
and
readily
similar
spec-
tral
features
were
found
(Table
1,
Figs.
S1
and
S2).
The
found
retention
values
as
well
are
not
very
far
from
that
of
Monolith
1a
(Fig.
5c).
Fig.
9a
shows
that
the
chromatographic
performance
of
Monolith
1c
is
reduced
when
compared
to
that
of
Monolith
1a
in
particular
in
the
right
hand
branch
of
the
plate
height
curve
and
for
the
retained
solutes.
This
is
seen
as
well
in
the
dependence

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
121
Fig.
9.
(a)
Plate
height
curves
from
Monolith
1c
(black
symbols)
in
contrast
to
Mono-
lith
1a
(gray
symbols)
with
symbol
assignments
to
solutes
as
in
Fig.
6;
(b)
plate
height
versus
retention
factor
at
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s
for
Monolith
1c
(black
filled
circles)
in
contrast
to
Monolith
1a
(gray
filled
circles).
Same
mobile
phase
as
in
Fig.
6.
of
performance
on
retention
at
constant
mobile
phase
velocities
(Fig.
9b).
Therefore,
we
can
assume
that
the
deterioration
in
per-
formance
comes
along
with
larger
structural
elements.
This
has
also
been
observed
for
conventional
polymer
based
monoliths
of
chemically
similar
polymer
material
but
increase
in
feature
size
[18,24].
This
situation
allows
direct
studies
of
specifics
of
phase
separation
and
quantitative
study
of
performance
according
to
a
varied
flow-through
pore
structure
at
otherwise
similar
retention
dynamics
and
fundamental
material’s
characteristics
in
this
new
class
of
materials.
3.4.6.
Impact
of
solute
identity,
mobile
phase
composition,
and
retention
on
performance
It
has
been
observed
recently
that
porous
polymer
monoliths
utilized
in
reversed-phase
[12,13,17,23],
as
well
as
in
hydrophilic
interaction
[25]
type
of
separations
show
a
strong
dependence
of
performance
on
retention.
This
is
true
within
a
homologous
series
of
structurally
similar
small
molecules
at
equivalent
mobile
phase
compositions,
as
well
as
for
a
given
solute
with
varied
mobile
phase
composition
(and
consequently
retention).
Such
potential
dependence
has
therefore
been
studied
with
Monolith
1a
that
showed
the
best
chromatographic
performance
in
the
separation
of
small
molecules
under
so
far
investigated
conditions.
In
order
to
assess
this
property,
we
performed
isocratic
elution
experiments
within
a
wide
range
of
mobile
phase
compositions
(from
30
to
Fig.
10.
(a)
Plot
of
retention
factor
measured
on
Monolith
1a
against
amount
of
acetonitrile
in
the
mobile
phase
(%,
v/v)
at
an
example
linear
chromatographic
flow
velocity
of
1.7
mm/s
with
the
same
symbol
assignments
to
solutes
as
in
Fig.
6;
(b)
variation
of
determined
plate
height
for
Monolith
1a
with
amount
of
acetonitrile
in
the
mobile
phase
(%,
v/v)
for
two
example
linear
chromatographic
flow
velocities
of
0.2
mm/s
and
1.7
mm/s.
Symbol
assignment
to
solutes
as
in
Fig.
6.
80%
acetonitrile
(v/v)
in
the
mobile
phase)
and
at
similar
linear
chromatographic
flow
velocities
of
0.2
and
1.7
mm/s.
Fig.
10a
shows
the
typical
reversed
phase
type
of
retention
behavior
for
Monolith
1a
with
varying
mobile
phase
composi-
tion
from
30
to
80%
acetonitrile
(v/v).
Retention
values
varied
from
0.12
to
23.5
within
this
plot.
Within
the
shown
range
of
mobile
phase
composition
we
find
little
impact
of
plate
heights
of
all
probe
solutes
at
a
linear
chromatographic
flow
velocity
of
0.2
mm/s
and
indistinguishable
from
the
non-retained
uracil
(Fig.
10b)
while
being
only
slightly
larger
than
the
non-retained
uracil
at
increased
flow
velocity.
At
first
glance,
this
indicates
the
substantially
different
chromatographic
behavior
when
compared
to
that
of
typical
polymer
monoliths.
This
will
be
detailed
in
the
discussion.
4.
Discussion
and
relation
of
chromatographic
behavior
to
that
of
state-of-the-art
polymer-
and
silica-based
monoliths
We
report
on
the
chromatographic
behavior
of
hybrid
inorganic-organic
porous
polymer
monoliths
prepared
by
a
radical-
mediated
step-growth
process.
Careful
choice
of
experimental
conditions
for
the
preparation
of
these
monoliths
identified

122
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
the
possibility
of
creation
of
near-ideal
step-growth
nanoscale
networks
under
certain
conditions
(Fig.
1c).
In
this
respect,
the
choice
of
the
thiol
monomers
and
its
effectiveness
in
linking
indi-
vidual
POSS
building
blocks
at
an
equimolar
stoichiometry
of
functional
groups
revealed
to
be
most
essential
criteria.
These
criteria
can
as
well
be
traced
by
here
presented
straightforward
experimental
methods.
Unlike
the
very
much
well-understood
free
radical
cross-
linking
(co)polymerization
processes
(conventionally
utilized
for
the
fabrication
of
polymer
monoliths)
that
is
typically
accom-
panied
by
a
compositional
drift
of
the
polymerization
mixture,
in
the
step-growth
cross-linking
(here
approached)
the
network-
forming
reaction
advances
in
a
stepwise
manner.
In
this
way
also,
the
constitution
of
resulting
nanoscale
networks
will
be
substantially
different.
A
network
generated
by
means
of
such
a
step-growth
process
will,
by
default,
be
more
uniform
in
terms
of
its
nanoscale
composition
and
physicochemical
properties
than
a
network
derived
by
a
free-radical
(co)polymerization
process.
We
hypothesize,
that
associated
to
the
former,
a
substantially
reduced
variation
in
cross-link
density
throughout
the
network
becomes
more
realistic
[21].
Eventually,
in
the
here
presented
process,
created
nuclei
in
ini-
tially
homogeneous
solution
will
phase
separate
at
a
certain
point
of
progression
of
the
step-growth
reaction.
Kinetics
of
polymeriza-
tion
and
thermodynamics
of
phase
separation
then
determine
the
microscale
structural
evolution
of
features
and,
consequently,
the
macroporous
structure
of
the
finally
created
monolithic
materials
(Fig.
3).
Practically,
we
have
assessed
the
macroporous
structure
by
elution
experiments
with
a
non-retained
tracer
(uracil)
and
found
with
this
information
at
hand,
the
overall
limit
in
perfor-
mance
for
the
present
set
of
columns.
The
identification
of
the
prime
origin
for
operation
ability
of
columns
in
reversed-phase
type
of
liquid
chromatographic
separations
identifies
as
well
an
improved
chromatographic
performance
of
both
monolith
types
when
the
column’s
phase
ratio
is
increased
(Figs.
4a,
6,
and
7).
This
is
indicated
by
a
more
effective
filling
of
capillary
cross-sectional
space
(Fig.
3a
and
b),
and
consequently
the
attainment
of
mono-
liths
with
apparently
less-pronounced
existence
of
microstructural
defects.
This
situation
leads
to
a
reduction
of
band
dispersion,
in
particular
the
Eddy
dispersion,
and
varying
development
of
a
“pseudo”
C-Term
deteriorating
performance
at
increased
flow
velocities
(Figs.
3
and
4).
A
detailed
study
of
associated
retention
dynamics
on
these
scaffolds
clearly
demonstrates
that
retention
processes
scale
with
the
overall
amount
of
hybrid
polymer
material
in
the
confine
(Fig.
5).
This
behavior
is
typical
of
chromatographic
stationary
phases
based
on
cross-linked
polymers,
in
particular
those
based
on
styrene/divinyl
benzene
chemistries
and
methacrylates
[18,23,24].
Though
we
are
dealing
with
more
densely
cross-linked
materi-
als,
this
indicates
accessibility
of
the
nanoscale
hybrid
polymer
gel
structure.
Based
on
experimental
results,
we
have
discussed
that
the
fun-
damental
constitution
of
the
cross-linked
material
can
have
three
possible
simplified
states
(Fig.
1c).
While
these
states
for
a
given
macroporous
constitution
of
the
monoliths
are
not
determining
for
the
performance
under
non-retained
elution
conditions
to
a
large
extent
(Fig.
4),
the
nanoscale
structure
(and
size
of
structural
fea-
tures)
of
the
monoliths
very
much
impact
the
efficiency
of
elution
for
retained
probe
solutes
that
undergo
partition
and
adsorption
processes
based
on
pure
physical
means.
While
the
associated
existence
of
gel
porosity
is
expected
still
being
present
based
on
rational
means
(Fig.
1c),
in
near-ideal
nanoscale
networks
(e.g.
Fig.
1c,
left
panel),
the
overall
system
may
show
a
more
homogeneous
energetic
distribution
of
partition
and
adsorption
sites.
Eventually,
this
leads
to
an
enhanced
performance
in
the
separation
of
small
retained
solutes,
very
close
to
that
of
the
non-retained
tracer
uracil
across
all
studied
mobile
phase
veloci-
ties,
retention
factors,
and
mobile
phase
compositions
(Figs.
6–10).
Such
aspect
becomes
supported
by
observation
of
poorer
perfor-
mance
for
retained
solutes
at
reduced
gel
fractions
when
using
the
EDDT
linker
(Figs.
1c,
middle
panel
and
7,
Table
1),
or
alterna-
tively
by
deliberately
introducing
more
variable
hole
free
volume
through
an
imbalance
in
stoichiometry
when
using
the
PETMP
linker
(Fig.
1c,
right
panel,
Fig.
8).
In
both
cases,
performance
dete-
riorates
due
to
the
presence
of
more
hole
free
volume
manifested
in
the
gel
porosity.
This
is
also
associated
with
a
clearly
observed
retention-dependent
performance
within
a
homologous
series
of
retained
solutes
(Figs.
7
and
8).
Enlarging
feature
size,
and
there-
fore
permeability,
for
a
given
monolithic
backbone
chemistry
as
well
showed
deterioration
of
performance
in
particular
for
retained
solutes
in
a
similar
retention
range
(Fig.
9).
With
near-ideal
nanoscale
networks
(Fig.
1c,
left
panel)
and
small
features,
mass
transfer
efficiency
is
best.
This
is
seen
in
per-
formance
of
retained
solutes
being
closest
to
that
of
non-retained
uracil
and
showing
similar
slopes
at
increased
flow
velocities
(Fig.
6).
Here
as
well
we
find
a
limited
influence
of
variable
mobile
phase
solvent
compositions
for
a
given
solute
(Fig.
10b).
SEM
images
of
the
column
also
show
the
most
esthetical
appearance
(Fig.
3a,
bottom
images).
Based
on
the
above
discussion,
we
clearly
identify
that
the
nanoscale
constitution
of
polymer
networks
determines
the
per-
formance
of
the
monoliths
for
small
retained
probe
solutes
that,
once
again,
does
not
require
a
special
set
of
dry-state
porous
prop-
erties
in
view
of
large
surface
areas
and
associated
population
of
mesopores
identified,
e.g.
by
nitrogen
sorption
analysis
(Table
1).
Therefore
also,
it
is
safe
to
assume
that
the
small
dry-state
surface
area
of
such
hybrid
polymers
is
not
judgmental
in
the
performance
of
elution
of
small
retained
solutes
since
sufficient
retention
and
selectivity
is
provided
by
all
of
these
columns
(Fig.
5).
In
fact,
the
apparently
acceptable
performance
for
non-retained
uracil
trans-
lates
to
a
very
good
performance
also
for
retained
solutes
under
conditions
in
which
the
backbone
nanoscale
structure
is
nearest-
to-ideal
(Fig.
1c,
left
panel,
Fig.
6).
The
practical
comparison
to
recent
work
on
polymer-
and
silica-
based
monoliths
published
[13],
provides
an
excellent
opportunity
for
comparison.
It
has
been
observed
that
efficiency–retention
rela-
tionships
vary
minimally
for
derivatized,
hierarchically
structured
silica-based
monoliths,
while
for
prime
examples
of
commer-
cialized
polymer
monoliths
efficiency-retention
relationships
vary
very
pronouncedly.
Polymer–eluent
interactions
have
been
made
responsible
for
such
variation.
This
is
associated
to
the
spatially
varying,
and
therefore
heterogeneous
gel
porosity
for
typical
poly-
mer
monoliths
[21].
As
a
result,
polymeric
monoliths
exhibit
variable
performances
depending
on
the
mobile
phase
composi-
tion
and
consequently
solute’s
retention,
further
amplified
by
local
variation
of
solvent
concentration,
since
the
polymer
favors
one
solvent
over
the
other
[21].
It
has
been
shown
that
such
effects
can
be
minimized,
for
example,
by
an
incomplete
polymerization
reac-
tion
at
which
the
compositional
drift
of
the
polymerization
mixture
has
not
build
up
significant
globule
scale
heterogeneities
[23,24].
To
find
overarching,
rational
means
of
how
the
current
set
of
columns
behave
in
relation
to
state-of-the-art
benchmark
polymer-
and
silica-based
monolithic
columns
as
well
as
capillary
scale
poly-
mer
monoliths,
we
have
utilized
a
recent
approach
of
normalized
plate
heights
[13].
For
this
purpose,
the
plate
heights
for
ben-
zyl
alcohol
and
ethyl
benzene,
as
retained
example
solutes,
were
divided
by
the
plate
height
of
non-retained
uracil
obtained
under
equivalent
chromatographic
conditions.
Then,
the
resulting
data
were
plotted
in
Fig.
11
together
with
data
found
in
the
literature
for
benchmark
polymer-
and
silica-based
monoliths
as
prime
exam-
ples
in
analytical
format
(from
Fig.
10a
and
b
of
ref.
[13]),
and
for
polymer-based
monoliths
in
capillary
format
(from
Fig.
10b
in
ref.

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
123
Fig.
11.
(a)
Normalized
plate
height
(the
plate
height
of
the
retained
solutes
(H)
was
divided
by
the
corresponding
plate
height
of
non-retained
uracil
(Huracil))
against
amount
of
acetonitrile
in
the
mobile
phase
(%,
v/v);
(b)
normalized
plate
height
against
retention
factor.
Data
obtained
in
this
study
at
a
linear
chromatographic
flow
velocity
of
0.5
mm/s
(blue,
orange,
and
green
symbols)
are
enriched
with
data
extracted
from
previous
studies
for
C18-derivatized
silica-based
(gray
symbols)
and
poly(styrene-co-
divinylbenzene)
monoliths
(black
symbols)
in
analytical
format
studied
in
ref.
[13].
Symbol
key:
benzyl
alcohol
(circles)
and
ethyl
benzene
(triangles);
Monolith
1a
(filled
blue
triangles
and
empty
blue
circles),
Monolith
1
(filled
green
triangle),
Monolith
1b
(half-filled
green
triangle),
Monolith
1c
(empty
green
triangle),
Monolith
2
(filled
orange
triangle)
and
Monolith
2a
(half-filled
orange
triangle).
In
(b)
the
data
corresponding
to
the
polymer-based
monoliths
in
capillary
format
(filled
red
triangles)
is
presented
at
increasing
polymerization
time
(increased
retention)
at
a
linear
chromatographic
flow
velocity
of
0.5
mm/s.
The
smallest
value
represents
a
polymerization
time
of
0.5
h
and
the
highest
of
3
h.
(For
interpretation
of
the
references
to
color
in
this
figure
legend,
the
reader
is
referred
to
the
web
version
of
this
article.)
[23]).
Though
the
chemistry
and
types
of
stationary
phases
may
physically
differ
substantially,
they
all
show
reversed-phase
type
of
retention
behavior,
allowing
for
a
very
practical
comparison.
Fig.
11a
and
b
shows
these
master
plots,
allowing
any
type
of
sta-
tionary
phase
qualitatively
be
set
in
context
in
overarching
terms
of
chromatographic
behavior.
We
can
clearly
see
that
all
hybrid
monoliths
prepared
in
this
study
(blue,
green,
and
orange
symbols)
behave
chromatographically
very
much
closer
to
the
reversed-
phase
silica-based
monolith
(gray
symbols).
Taking
as
an
example
Monolith
1a
(empty
blue
circles
for
benzyl
alcohol
and
filled
blue
triangles
for
ethyl
benzene),
we
identify
behavior
likewise
the
C-
18
derivatized
silica-based
example
(see
zoom
area
of
Fig.
11a).
In
fact,
this
material
shows
the
most
ideal
chromatographic
behav-
ior
demonstrated
in
this
study.
We
have
also
shown
for
Monolith
1a,
that
these
data
are
expected
insensitive
to
varying
linear
chro-
matographic
flow
velocities
as
well
since
the
plate
height
curves
collapse
to
one
single
master
curve
(black
symbols
in
Fig.
6a).
Monolith
1
(filled
green
triangles)
and
Monolith
1a
(empty
blue
circles
for
benzyl
alcohol
and
filled
blue
triangles
for
ethyl
ben-
zene)
showed
the
lowest
variations
in
plate
height
values
under
non-retained
and
retained
elution
conditions.
These
monoliths
also
possess
the
nearest-to-ideal
nanoscale
network
structures
(highest
gel
fractions
seen
in
Table
1,
Fig.
1c,
left
panel).
In
the
zoom
area
of
Fig.
11a,
we
can
observe
that
highest
normalized
plate
heights
at
the
same
example
mobile
phase
com-
position
(50%
acetonitrile/water
(v/v))
were
found
for
Monolith
1b
(half-filled
green
triangles),
followed
by
Monolith
1c
(empty
green
triangles),
followed
by
Monoliths
2
and
2a
(filled
orange
triangles
and
half-filled
orange
triangles).
These
observations
substantialize
the
fact
that
near-ideal
nanoscale
networks
(Fig.
1c,
left
panel)
as
well
as
small
structural
features
(Fig.
3a,
lower
images)
contribute
to
an
ideal
chromatographic
behavior,
as
most
closely
realized
with
Monolith
1a
(empty
blue
circles,
filled
blue
triangles).
It
is
therefore
safe
to
assume,
that
indeed
the
chromatographic
behavior
contrasts
to
the
analytical
polymer-based
ones
(black
symbols),
i.e.
where
performance
is
very
much
smaller
for
retained
small
solutes
than
for
the
non-retained
uracil,
as
well
being
varied
with
mobile
phase
solvent
composition
(Fig.
11a),
and
conse-
quently
with
retention
factors
(Fig.
11b).
To
establish
a
final
link
between
the
chromatographic
behav-
ior
of
the
capillary-scale
hybrid
polymer
monoliths
of
this
study,
discussed
analytical-scale
polymer-
and
silica
-based
monoliths
to

124
F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
other
recent
work,
we
as
well
report
on
data
of
standard
methacry-
late
based
stationary
phases
prepared
under
variable
conversion
of
monomeric
precursors
in
the
capillary
format
[23].
Here
an
enhanced
chromatographic
performance
was
found
at
an
incom-
plete
conversion.
This
study
as
well
showed
that
at
an
increased
conversion
of
monomeric
precursors,
retention
of
a
homologous
series
of
alkyl
benzenes
increased.
This
increase
in
retention
was
accompanied
by
a
deterioration
in
chromatographic
performance
explained
by
the
pronounced
development
of
a
cross-link
density
distribution
in
larger
structural
features
constituting
the
backbone
of
the
monolith.
The
origin
of
this
phenomenon
was
explained
based
on
a
compositional
drift
in
the
polymerization
mixture
during
free-radical
cross-linking
copolymerization
[23].
Red
filled
triangles
in
Fig.
11b
represent
data
of
ethyl
benzene
from
such
monoliths
derived
the
same
way
from
available
literature
as
other
data
reported
in
Fig.
11.
In
fact,
these
data
demonstrate
that
the
performance
for
ethylbenzene
can
come
very
close
to
that
of
the
non-retained
uracil
however,
as
well
enters
the
range
of
perfor-
mance
found
for
typical
polymer
monoliths,
i.e.
very
much
alike
commercial
examples.
This
somewhat
validates
that
our
approach
of
normalized
plate
heights
is
a
very
useful
means
to
identify
over-
all
chromatographic
behavior
and
its
deviation
from
ideality
for
almost
any
type
of
stationary
phase.
5.
Conclusions
and
incentives
for
further
research
We
have
demonstrated
and
explained
how
a
principally
new
class
of
hybrid
monolithic
materials
via
a
radical-mediated
step
growth
implementation
can
be
created.
While
the
choice
of
chem-
istry
applicable
reaches
beyond
the
one
demonstrated
here,
fine
control
of
experimental
synthetic
conditions
and
careful
analysis
of
resultant
monolithic
structures
with
a
suitable
suite
of
com-
plementary
techniques
allowed
for
identification
of
the
origin
of
an
improved
performance
in
liquid
chromatography
for
the
first
time.
Clearly,
the
creation
of
near-ideal
nanoscale
networks
as
emanating
origin
for
the
resultant
monolithic
structures
is
the
key
to
success
for
creation
of
efficient
stationary
phases.
In
par-
ticular,
a
homogeneous
energetic
distribution
of
partition
and
adsorption
sites
as
a
key
for
retention,
was
identified
to
lead
to
an
enhanced
performance
in
the
separation
of
small
retained
solutes.
This
study
as
well
shows
that
a
careful
consideration
of
the
material
formation
concept
can
provide
the
monoliths
with
a
good
performance,
in
particular
under
retained
elution
conditions.
Associated
guidelines
for
hybrid
monolith
preparation
established
in
the
present
study
are
the
following
in
order
of
importance:
(i)
choice
of
thiol-linker
creating
near-ideal
nanoscale
networks
with
vinylPOSS,
(ii)
stoichiometry
of
complementary
reactive
func-
tional
groups
that
imparts
control
over
occurrence
of
nanoscale
network
defects
and
clearly
observed
retention-(in)dependent
per-
formance,
as
well
as
(iii)
porogenic
solvent
composition
impacting
topology
of
the
flow-through
pore
space.
Ongoing
experiments
shall
also
address
a
direct
nanoscale
assessment
of
the
backbone
network
structures
and
poten-
tial
microscale
heterogeneities
with
recent
approaches
at
hand
[21,22,53,54].
A
necessary
add-on
to
understand
the
solvated
state
properties
of
the
materials
may
as
well
be
size
exclusion
chro-
matography
of
the
here
utilized
probe
solutes
resolving
potential
differences
in
the
nanometer-sized
pore
space
[18].
In
as
much
as
the
present
study
concerns
small
molecules,
further
work
is
nec-
essary
to
address
the
performance
in
the
separation
of
proteins.
Here
the
presence
of
a
(grafted)
gel
structure
of
low
cross-linking
and
the
associated
nanoscale
solvation
is
known
to
impact
protein
binding
and
transport
as
well
as
dynamics
of
non-equilibrium
(i.e.
gradient)
chromatographic
performance
[19].
While
the
overall
performance
of
this
new
step-growth
hybrid
monoliths
do
not
yet
reach
the
performance
of
the
silica-based
ones,
they
possess
such
principal
potential.
As
it
was
discussed
in
terms
of
retention-sensitive
performance,
the
monoliths’
chro-
matographic
behavior
under
thermodynamically
relevant
elution
conditions
is
by
far
closer
to
the
derivatized
silica-based
materi-
als
and
clearly
contrasts
their
benchmark
polymeric
counterparts
based
on
free-radical
cross-linking
copolymerization
as
well
as
typical
capillary
polymer
monoliths
derived
from
such
processes.
With
this
opportunity
at
hand,
we
note
that
the
column
efficiency
for
all
solutes
deteriorates
at
increased
flow
velocities,
an
aspect
we
believe
of
having
its
origin
in
a
still
persistent
heterogeneous
flow-through
pore
structure,
in
particular
across
the
columns’
cross-section.
This
phenomenon
is
expected
to
have
its
origin
in
the
kinetics
of
polymerization
and
thermodynamics
of
phase
sepa-
ration.
Though
this
aspect
is
known
to
possibly
result
in
a
“pseudo”
C-Term,
the
here
presented
techniques
do
not
allow
its
resolution
and
substantiation.
However,
the
present
work
demonstrates
that
identification
of
such
improvements,
potentially
made
possible
by
particularly
addressing
phase
separation
and
microscale
structure
evolution,
might
be
a
promising
objective
in
future
studies.
We
will
leave
such
improvement
to
forthcoming
investigations
that
are
likely
to
be
reported
soon.
Acknowledgements
This
work
was
supported
by
the
Austrian
Science
Fund
(FWF)
under
project
number
[P24557-N19].
We
would
like
to
thank
Gün-
ther
Hesser
at
ZONA
(Johannes
Kepler
University)
for
help
with
the
scanning
electron
microscopy
(SEM)
measurements
and
Marcin
Pawliczek
at
RECENDT
GmbH
for
allowing
us
to
use
the
portable
Raman
spectrometer.
Appendix
A.
Supplementary
data
Supplementary
data
associated
with
this
article
can
be
found,
in
the
online
version,
at
http://dx.doi.org/10.1016/j.chroma.2015.08.
019
References
[1]
P.
Jandera,
Stationary
and
mobile
phases
in
hydrophilic
interaction
chromatog-
raphy:
a
review,
Anal.
Chim.
Acta
692
(2011)
1–25.
[2]
L.
Nováková,
L.
Havlíková,
H.
Vlˇ
cková,
Hydrophilic
interaction
chromatogra-
phy
of
polar
and
ionizable
compounds
by
UHPLC,
TrAC:
Trends
Anal.
Chem.
63
(2014)
55–64.
[3]
A.
Vailaya,
C.
Horvath,
Retention
in
reversed-phase
chromatography:
partition
or
adsorption?
J.
Chromatogr.
A
829
(1998)
1–27.
[4]
L.
Zhang,
J.L.
Rafferty,
J.I.
Siepmann,
M.R.
Schure,
How
reversed-phase
liquid
chromatography
works,
LCGC
N.
Am.
31
(2014)
1.
[5]
F.
Gritti,
A.
Höltzel,
U.
Tallarek,
G.
Guiochon,
The
relative
importance
of
the
adsorption
and
partitioning
mechanisms
in
hydrophilic
interaction
liquid
chro-
matography,
J.
Chromatogr.
A
1376
(2015)
112–125.
[6]
S.
Hjertén,
J.-L.
Liao,
R.
Zhang,
High-performance
liquid
chromatography
on
continuous
polymer
beds,
J.
Chromatogr.
473
(1989)
273–275.
[7]
F. ˇ
Svec,
J.M.J.
Fréchet,
Continuous
rods
of
macroporous
polymer
as
high-
performance
liquid
chromatography
separation
media,
Anal.
Chem.
64
(1992)
820–822.
[8]
Q.C.
Wang,
F. ˇ
Svec,
J.M.J.
Fréchet,
Macroporous
polymeric
stationary-phase
rod
as
continuous
separation
medium
for
reversed-phase
chromatography,
Anal.
Chem.
65
(1993)
2243–2248.
[9]
Q.C.
Wang,
F. ˇ
Svec,
J.M.J.
Fréchet,
Reversed-phase
chromatography
of
small
molecules
and
peptides
on
a
continuous
rod
of
macroporous
poly
(styrene-co-
divinylbenzene),
J.
Chromatogr.
A
669
(1994)
230–235.
[10]
H.
Minakuchi,
K.
Nakanishi,
N.
Soga,
N.
Ishizuka,
N.
Tanaka,
Octadecylsilylated
porous
silica
rods
as
separation
media
for
reversed-phase
liquid
chromatogra-
phy,
Anal.
Chem.
68
(1996)
3498–3501.
[11]
K.
Nakanishi,
H.
Minakuchi,
N.
Soga,
N.
Tanaka,
Double
pore
silica
gel
monolith
applied
to
liquid
chromatography,
J.
Sol-Gel
Sci.
Technol.
8
(1997)
547–552.
[12]
I.
Nischang,
Porous
polymer
monoliths:
morphology,
porous
properties,
poly-
mer
nanoscale
gel
structure
and
their
impact
on
chromatographic
performance,
J.
Chromatogr.
A
1287
(2013)
39–58.

F.
Alves,
I.
Nischang
/
J.
Chromatogr.
A
1412
(2015)
112–125
125
[13]
T.J.
Causon,
I.
Nischang,
Critical
differences
in
chromatographic
proper-
ties
of
silica-
and
polymer-based
monoliths,
J.
Chromatogr.
A
1358
(2014)
165–171.
[14]
J.L.
Rafferty,
L.
Zhang,
J.I.
Siepmann,
M.R.
Schure,
Retention
mechanism
in
reversed-phase
liquid
chromatography:
a
molecular
perspective,
Anal.
Chem.
79
(2007)
6551–6558.
[15]
L.
Rafferty,
J.I.
Siepmann,
M.R.
Schure,
in:
E.
Grushka,
N.
Grinberg
(Eds.),
Advances
in
Chromatography,
vol.
48,
CRC
Press,
Boca
Raton,
FL,
2009,
pp.
1–53.
[16]
R.K.
Lindsey,
J.L.
Rafferty,
B.L.
Eggimann,
J.I.
Siepmann,
M.R.
Schure,
Molecular
simulation
studies
of
reversed-phase
liquid
chromatography,
J.
Chromatogr.
A
1287
(2013)
60–82.
[17]
I.
Nischang,
I.
Teasdale,
O.
Bruggemann,
Porous
polymer
monoliths
for
small
molecule
separations:
advancements
and
limitations,
Anal.
Bioanal.
Chem.
400
(2011)
2289–2304.
[18]
I.
Nischang,
On
the
chromatographic
efficiency
of
analytical
scale
column
for-
mat
porous
polymer
monoliths:
interplay
of
morphology
and
nanoscale
gel
porosity,
J.
Chromatogr.
A
1236
(2012)
152–163.
[19]
I.
Nischang,
Impact
of
biomolecule
solute
size
on
the
transport
and
performance
characteristics
of
analytical
porous
polymer
monoliths,
J.
Chromatogr.
A
1354
(2014)
56–64.
[20]
T.J.
Causon,
E.F.
Hilder,
I.
Nischang,
Impact
of
mobile
phase
composition
on
the
performance
of
porous
polymeric
monoliths
in
the
elution
of
small
molecules,
J.
Chromatogr.
A
1263
(2012)
108–112.
[21]
M.
Laher,
T.J.
Causon,
W.
Buchberger,
S.
Hild,
I.
Nischang,
Assessing
the
nanoscale
structure
and
mechanical
properties
of
polymer
monoliths
used
for
chromatography,
Anal.
Chem.
85
(2013)
5645–5649.
[22]
I.
Nischang,
T.J.
Causon,
Porous
polymer
monoliths:
from
their
fundamental
structure
to
opportunities
for
analytical
engineering
applications,
TrAC:
Trends
Anal.
Chem.
(2015)
(in
press).
[23]
I.
Nischang,
O.
Bruggemann,
On
the
separation
of
small
molecules
by
means
of
nano-liquid
chromatography
with
methacrylate-based
macroporous
polymer
monoliths,
J.
Chromatogr.
A
1217
(2010)
5389–5397.
[24]
I.
Nischang,
I.
Teasdale,
O.
Bruggemann,
Towards
porous
polymer
monoliths
for
the
efficient,
retention-independent
performance
in
the
isocratic
separation
of
small
molecules
by
means
of
nano-liquid
chromatography,
J.
Chromatogr.
A
1217
(2010)
7514–7522.
[25]
C¸
.
Kip,
D.
Erkakan,
A.
Gökaltun,
B.
C¸
elebi,
A.
Tuncel,
Synthesis
of
a
reactive
polymethacrylate
capillary
monolith
and
its
use
as
a
starting
material
for
the
preparation
of
a
stationary
phase
for
hydrophilic
interaction
chromatography,
J.
Chromatogr.
A
1396
(2015)
86–97.
[26]
C.E.
Hoyle,
C.N.
Bowman,
Thiol-ene
click
chemistry,
Angew.
Chem.
Int.
Ed.
49
(2010)
1540–1573.
[27]
M.-L.
Chen,
J.
Zhang,
Z.
Zhang,
B.-F.
Yuan,
Q.-W.
Yu,
Y.-Q.
Feng,
Facile
preparation
of
organic-silica
hybrid
monolith
for
capillary
hydrophilic
liquid
chromatog-
raphy
based
on
“thiol-ene”
click
chemistry,
J.
Chromatogr.
A
1284
(2013)
118–125.
[28]
Z.
Lin,
X.
Tan,
R.
Yu,
J.
Lin,
X.
Yin,
L.
Zhang,
H.
Yang,
One-pot
preparation
of
glutathione-silica
hybrid
monolith
for
mixed-mode
capillary
liquid
chro-
matography
based
on
“thiol-ene”
click
chemistry,
J.
Chromatogr.
A
1355
(2014)
228–237.
[29]
S.
Shen,
F.
Ye,
C.
Zhang,
Y.
Xiong,
L.
Su,
S.
Zhao,
Preparation
of
polyhedral
oligomeric
silsesquioxane
based
hybrid
monoliths
by
thiol-ene
click
chemistry
for
capillary
liquid
chromatography,
Analyst
140
(2015)
265–271.
[30]
F.
Alves,
I.
Nischang,
Tailor-made
hybrid
organic–inorganic
porous
materials
based
on
polyhedral
oligomeric
silsesquioxanes
(POSS)
by
the
step-
growth
mechanism
of
thiol-ene
“click”
chemistry,
Chem.
Eur.
J.
19
(2013)
17310–17313.
[31]
F.
Alves,
I.
Nischang,
A
simple
approach
to
hybrid
inorganic–organic
step-
growth
hydrogels
with
scalable
control
of
physicochemical
properties
and
biodegradability,
Polym.
Chem.
6
(2015)
2183–2187.
[32]
H.
Lin,
J.J.
Ou,
Z.S.
Liu,
H.W.
Wang,
J.
Dong,
H.F.
Zou,
Thiol-epoxy
click
polymerization
for
preparation
of
polymeric
monoliths
with
well-defined
3D
framework
for
capillary
liquid
chromatography,
Anal.
Chem.
87
(2015)
3476–3483.
[33]
H.
Lin,
J.J.
Ou,
Z.S.
Liu,
H.W.
Wang,
J.
Dong,
H.F.
Zou,
Facile
construction
of
macro-
porous
hybrid
monoliths
via
thiol-methacrylate
Michael
addition
click
reaction
for
capillary
liquid
chromatography,
J.
Chromatogr.
A
1379
(2015)
34–42.
[34]
H.
Lin,
J.J.
Ou,
Z.B.
Zhang,
J.
Dong,
H.F.
Zou,
Ring-opening
polymerization
reaction
of
polyhedral
oligomeric
silsesquioxanes
(POSSs)
for
preparation
of
well-controlled
3D
skeletal
hybrid
monoliths,
Chem.
Commun.
49
(2013)
231–233.
[35]
Z.S.
Liu,
J.J.
Ou,
H.
Lin,
Z.Y.
Liu,
H.W.
Wang,
J.
Dong,
H.F.
Zou,
Photoinduced
thiol-ene
polymerization
reaction
for
fast
preparation
of
macroporous
hybrid
monoliths
and
their
application
in
capillary
liquid
chromatography,
Chem.
Commun.
50
(2014)
9288–9290.
[36]
Z.S.
Liu,
J.J.
Ou,
H.
Lin,
H.W.
Wang,
Z.Y.
Liu,
J.
Dong,
H.F.
Zou,
Preparation
of
monolithic
polymer
columns
with
homogeneous
structure
via
photoinitiated
thiol-yne
click
polymerization
and
their
application
in
separation
of
small
molecules,
Anal.
Chem.
86
(2014)
12334–12340.
[37]
F.
Alves,
P.
Scholder,
I.
Nischang,
Conceptual
design
of
large
surface
area
porous
polymeric
hybrid
media
based
on
polyhedral
oligomeric
silsesquiox-
ane
precursors:
preparation,
tailoring
of
porous
properties,
and
internal
surface
functionalization,
ACS
Appl.
Mater.
Interfaces
5
(2013)
2517–2526.
[38]
I.
Nischang,
F.
Svec,
J.M.J.
Frechet,
Downscaling
limits
and
confinement
effects
in
the
miniaturization
of
porous
polymer
monoliths
in
narrow
bore
capillaries,
Anal.
Chem.
81
(2009)
7390–7396.
[39]
C.E.
Hoyle,
A.B.
Lowe,
C.N.
Bowman,
Thiol-click
chemistry:
a
multifaceted
tool-
box
for
small
molecule
and
polymer
synthesis,
Chem.
Soc.
Rev.
39
(2010)
1355–1387.
[40]
I.
Nischang,
O.
Brueggemann,
F. ˇ
Svec,
Advances
in
the
preparation
of
porous
polymer
monoliths
in
capillaries
and
microfluidic
chips
with
focus
on
mor-
phological
aspects,
Anal.
Bioanal.
Chem.
397
(2010)
953–960.
[41]
O.
Nunez,
K.N.
Nakanishi,
Tanaka,
Preparation
of
monolithic
silica
columns
for
high-performance
liquid
chromatography,
J.
Chromatogr.
A
1191
(2008)
231–252.
[42]
K.
Hormann,
T.
Müllner,
S.
Bruns,
A.
Höltzel,
U.
Tallarek,
Morphology
and
separation
efficiency
of
a
new
generation
of
analytical
silica
monoliths,
J.
Chro-
matogr.
A
1222
(2012)
46–58.
[43]
R.
Meinusch,
K.
Hormann,
R.
Hakim,
U.
Tallarek,
B.M.
Smarsly,
Synthesis
and
morphological
characterization
of
phenyl-modified
macroporous–mesoporous
hybrid
silica
monoliths,
RSC
Advances
5
(2015)
20283–20294.
[44]
T.
Hara,
H.
Kobayashi,
T.
Ikegami,
K.
Nakanishi,
N.
Tanaka,
Performance
of
monolithic
silica
capillary
columns
with
increased
phase
ratios
and
small-sized
domains,
Anal.
Chem.
78
(2006)
7632–7642.
[45]
K.N.
Smirnov,
I.A.
Dyatchkov,
M.V.
Telnov,
A.V.
Pirogov,
O.A.
Shpigun,
Effect
of
monomer
mixture
composition
on
structure
and
chromatographic
properties
of
poly(divinylbenzene-co-ethylvinylbenzene-co-2-hydroxyethyl
methacry-
late)
monolithic
rod
columns
for
separation
of
small
molecules,
J.
Chromatogr.
A
1218
(2011)
5010–5019.
[46]
F.
Gritti,
G.
Guiochon,
Mass
transport
of
small
retained
molecules
in
polymer-
based
monolithic
columns,
J.
Chromatogr.
A
1362
(2014)
49–61.
[47]
S.-L.
Lin,
Y.-R.
Wu,
T.-Y.
Lin,
M.-R.
Fuh,
Preparation
and
evaluation
of
poly(alkyl
methacrylate-co-methacrylic
acid-co-ethylene
dimethacrylate)
monolithic
columns
for
separating
polar
small
molecules
by
capillary
liquid
chromatogra-
phy,
Anal.
Chim.
Acta
871
(2015)
57–65.
[48]
J.
Wang,
L.
Bai,
Z.
Wei,
J.
Qin,
Y.
Ma,
H.
Liu,
Incorporation
of
ionic
liquid
into
porous
polymer
monoliths
to
enhance
the
separation
of
small
molecules
in
reversed-phase
high-performance
liquid
chromatography,
J.
Sep.
Sci.
(2015)
2101–2108.
[49]
A.
Laaniste,
A.
Kruve,
I.
Leito,
Ensuring
repeatability
and
robustness
of
poly(glycidyl
methacrylate-co-ethylene
dimethacrylate)
HPLC
monolithic
columns
of
3
mm
id
through
covalent
bonding
to
the
column
wall,
J.
Sep.
Sci.
36
(2013)
2458–2463.
[50]
A.M.
Siouffi,
About
the
C
term
in
the
van
Deemter’s
equation
of
plate
height
in
monoliths,
J.
Chromatogr.
A
1126
(2006)
86–94.
[51]
D.
Hlushkou,
S.
Bruns,
A.
Höltzel,
U.
Tallarek,
From
pore
scale
to
column
scale
dispersion
in
capillary
silica
monoliths,
Anal.
Chem.
82
(2010)
7150–7159.
[52]
S.
Bruns,
J.P.
Grinias,
L.E.
Blue,
J.W.
Jorgenson,
U.
Tallarek,
Morphology
and
separation
efficiency
of
low-aspect-ratio
capillary
ultrahigh
pressure
liquid
chromatography
columns,
Anal.
Chem.
84
(2012)
4496–4503.
[53]
T.
Müllner,
A.
Zankel,
C.
Mayrhofer,
H.
Reingruber,
A.
Höltzel,
Y.
Lv,
F.
Svec,
U.
Tallarek,
Reconstruction
and
characterization
of
a
polymer-based
mono-
lithic
stationary
phase
using
serial
block-face
scanning
electron
microscopy,
Langmuir
28
(2012)
16733–16737.
[54]
P.
Aggarwal,
V.
Asthana,
J.S.
Lawson,
H.D.
Tolley,
D.R.
Wheeler,
B.A.
Mazzeo,
M.L.
Lee,
Correlation
of
chromatographic
performance
with
morphological
features
of
organic
polymer
monoliths,
J.
Chromatogr.
A
1334
(2014)
20–29.