Interfacial and Nanoconfinement Effects Decrease the Excited-State Acidity of Polymer-Bound Photoacids

Article

Interfacial and Nanoconﬁnement Effects

Decrease the Excited-State Acidity of Polymer-

Bound Photoacids

Christopher D. Sanborn, Jenu

Varghese Chacko, Michelle

Digman, Shane Ardo

ardo@uci.edu

HIGHLIGHTS

Binding photoacids to conical

nanopores decreases their acidity

When photoacids are bound to

nanopores they exhibit less

excited-state proton transfer

The acidity and density of

functional groups in single

nanopores is quantiﬁed

We previously reported photovoltaic action from photoacid-dye-modiﬁed ion-

exchange membranes. A more controlled model system for those materials are

photoacid-modiﬁed nanopores in poly(ethylene terephthalate) reported herein.

Photoacids bound to the sub-10-nm-sized tips of these nanoporous poly(ethylene

terephthalate) materials exhibited decreases in ground-state and excited-state

acidity versus the same photoacid dyes dissolved in solution. The data indicate

that nano-conﬁnement and local electrostatics are important considerations when

designing light-to-ionic energy conversion devices with possible applications in

energy conversion and neuron triggering.

Sanborn et al., Chem 5, 1648–1670

June 13, 2019 ª 2019 Published by Elsevier Inc.

https://doi.org/10.1016/j.chempr.2019.04.022

Article

Interfacial and Nanoconﬁnement Effects

Decrease the Excited-State

Acidity of Polymer-Bound Photoacids

Christopher D. Sanborn,¹Jenu Varghese Chacko,²Michelle Digman,²and Shane Ardo^1,3,4,5,

*

SUMMARY

The Bigger Picture

A major challenge facing

Photo-initiated ion transport on the nanoscale is relevant to various ﬁelds

including energy conversion, neuron triggering, and biomimetic processes. To

study this phenomenon, we synthesized and evaluated photoacid-modiﬁed

polymers, which generate hydrated protons in response to visible-light excita-

tion. Pyrenol photoacid dye molecules were covalently bonded within conical

nanopores that had been track etched in poly(ethylene terephthalate). The

data suggest that ꢀ90% of the nanopore surfaces were modiﬁed with photoa-

cids and that photoacids were, on average, bound through three sulfonamide

groups. In comparison to photoacids dissolved in an aqueous solution, photoa-

cids bound speciﬁcally to the nanoporous tip region were, on average, weaker

acids in their ground and excited states, which we presume is due to differences

in surface potential or solvation environment in the conﬁned nanopores. These

results suggest that ion transport initiated by nanoconﬁned photoacids will

require careful molecular engineering to enable efﬁcient light-to-ionic energy

conversion.

humanity is the availability of

clean and potable water for

human consumption and

agriculture. Over time, this

problem is expected to worsen

because of population growth and

increases in the severity and

frequency of global climate

disruption. As such, new

technologies for low-cost,

efﬁcient, and distributed

desalination of salt water are

desired. Toward this, our group

recently reported ion-selective

membranes that directly pump

ions when illuminated with

sunlight. A critical component in

the technology is the polymer-

bound photoacid dye, which is

carefully characterized herein. In

our initial materials systems,

controlled quantitative

INTRODUCTION

Control of the direction and timing of energy and charge transport on the nano-

scale and macroscale has allowed for various innovations, from advanced micro-

electronics and photovoltaics to life. While in most artiﬁcial systems energy and

charge are transported using electronic processes,¹nature on the other hand

also uses ionic processes.^2,3For example, action potentials are transmitted along

neurons via a series of orchestrated ion ﬂuxes across lipid bilayers,⁴and the

photosynthetic processes in photosystem II of green plants involve orthogonal

transport of electrons and protons.⁵In each of these biological systems, an asym-

metry across the membrane leads to a preferential direction for species transport

often based on the sign of their charge^6–9and which is manifest as current

rectiﬁcation.

measurements of the properties

of polymer-bound photoacids

were extremely challenging.

Using the knowledge gained from

studies herein, we are hopeful that

the efﬁciency of our technology

can be increased rapidly, to a

point where it enables

Several artiﬁcial systems have also been shown to rectify ionic current: ion-ex-

change membranes,^10–14nanopores,^15–26and nanopipettes.^27–32Utilizing the ma-

terials asymmetry that results in this current rectiﬁcation, light-driven proton pumps

have been demonstrated.^33–36However, until recently,^37–39none of these materials

utilized covalently bound photoacid-dye molecules.^40,41Photoacids enable direct

conversion of light energy into a change in electrochemical potential of protons

and are therefore well-suited for light-driven proton pumping. Photon absorption

by a photoacid dye results in a decrease in the pK_aof one of its protic functional

groups, therefore making it more acidic in its excited state.⁴²This results in net

inexpensive and autonomous

desalination of distributed

saltwater sources.

1648 Chem 5, 1648–1670, June 13, 2019 ª 2019 Published by Elsevier Inc.

liberation of a proton charge carrier that can then be transported in a direction that

is dictated by the electrostatic asymmetry of the material to which it is bound. If the

built-in asymmetry of the material results in ionic current rectiﬁcation due to the

presence of selective contacts on each side of the material,⁴³light should impart

vectorial pumping of ions. Recently, we demonstrated that photoacid-dye-modiﬁed

Naﬁon ion-exchange membranes exhibit photovoltaic action when illuminated with

visible light.^37,39Absorption of light by Naﬁon-bound photoacids resulted in a

change in the local concentration of protons,^44,45thus perturbing the chemical

speciation within the hydrated charged nanopores of Naﬁon away from its initial

state and generating a small photovoltage. The observation of a photocurrent

was consistent with directional ion transport of the photo-liberated protons

driven by an asymmetry in the hydrated charged nanopores of Naﬁon. The

mechanistic causes for this small photovoltage and small photocurrent from

unoptimized photoacid-dye-modiﬁed Naﬁon are not well understood. This

prompted us to further study the fundamental photophysical properties of

photoacid-dye molecules bound to hydrated charged nanoporous polymers, such

as Naﬁon and poly(ethylene terephthalate) (PET).

Herein, we utilize charged polymer nanopores as model systems for ion-exchange

membranes to further interrogate the photovoltaic ion-pumping process previ-

ously observed in photoacid-dye-modiﬁed Naﬁon. Nanopores are excellent

models for ion-exchange membranes, like Naﬁon, because Naﬁon phase segre-

gates into micelle-like hydrophilic regions that are spaced on the order of several

nanometers and bear negatively charge covalently bound (ﬁxed) sulfonate groups

on their interior.^46–48The model systems that we utilize herein consist of ion-recti-

fying track-etched conical nanopores in PET with tip diameters as small as several

nanometers and bearing negatively charged ﬁxed carboxylate groups on their

interior. Through careful characterization of these model nanoporous rectifying

materials functionalized with custom photoacid molecules, we further elucidated

the role that conﬁnement plays on the photophysical properties of the bound pho-

toacid-dye molecules. Our results may help explain some of the limitations in the

light-driven ion pumping process that we observed using photoacid-dye-modiﬁed

Naﬁon membranes.

RESULTS AND DISCUSSION

Molecular Synthesis of a Bondable Photoacid Molecule

Custom photoacid, tris(sulfonamide)pyrenol 1, was synthesized from pyranine using

a four-step synthetic protocol (Scheme 1). Pyranine was treated with acetic anhy-

dride under alkaline conditions to afford an acetyl-protected hydroxyl group fol-

lowed by reﬂuxing in excess thionyl chloride with catalytic N,N-dimethylformamide

to activate the three sulfonate groups as sulfonyl chlorides. In a solution of 2:7:1

THF:CH₃CN:H₂O (v/v/v), two equivalents of N-Boc-ethylenediamine per sulfonyl

chloride were added such that one equivalent formed the sulfonamide and HCl,

which was neutralized by the other equivalent. The product was then deacetylated

via addition of NaOH, and the Boc protecting group was removed via triﬂuoroacetic

acid treatment to yield 1. Additional puriﬁcation was achieved using reverse-phase

high-performance liquid chromatography.

¹Department of Chemistry, University of

California, Irvine, Irvine, CA 92697, USA

²Department of Biomedical Engineering,

University of California, Irvine, Irvine, CA 92697,

USA

³Department of Materials Science & Engineering,

University of California, Irvine, Irvine, CA 92697,

USA

⁴Department of Chemical & Biomolecular

Engineering, University of California, Irvine, Irvine,

CA 92697, USA

Measurement of Acid Dissociation Constants for Photoacids in Solution

The most acidic pK_avalue of 1 dissolved in aqueous buffered solution was deter-

mined to be 5.80 G 0.03 via an acid–base titration procedure with spectrophoto-

metric detection and non-linear least-squares ﬁtting of the data to the Hill equation

⁵Lead Contact

*Correspondence: ardo@uci.edu

https://doi.org/10.1016/j.chempr.2019.04.022

Chem 5, 1648–1670, June 13, 2019 1649

Scheme 1. Photoacid Synthetic Protocol

Synthesis of 1 from pyranine by sulfonate activation.

See also Figure S6.

Measurement of Ground-State Acidities of Polymer-Bound Photoacids

For photoacid-functionalized nanopores to serve as efﬁcient light-driven ion pumps,

PET_n/1 should be a weak acid in its ground state and a strong photoacid in its excited

state. For PET₈/1 immersed in an aqueous buffered solution, pK_a= 5.1 G 0.2 for the

hydroxyl group of 1 as determined by an acid–base titration procedure with spectro-

photometric detection (Figure 4). This value is 0.7 pH units more acidic than the pK_a

of the hydroxyl group of 1 when dissolved in solution but is consistent with the pK_a

value of ꢀ5.4 observed for a model of PET_n/1, i.e., tris(methylterephthalate)-capped

1 (photoacid 2), in aqueous solution containing no intentional inert electrolyte salt

(see Figure S10; Scheme S1). These data support that terephthalate polymer units

stabilize the conjugate base of 1, which is similar to the stabilization effects lyso-

zymes have on 8-hydroxypyrene-1,3,6-trisulfonate via hydrogen bonding, van der

Waals interactions, and electrostatic interactions.⁵¹The large degree of non-ideality

observed in the acid–base titration data for PET₈/1 (jnj = 0.55–0.60) is most likely

caused by a distribution of pore sizes and/or nanoenvironments that are not present

when 1 is dissolved in solution.

The acid–base titration procedure with spectrophotometric detection reported on

the average pK_aof all photoacid molecules bound to the PET surface but did not

provide information about the pK_aof photoacids bound speciﬁcally in the tips of

the nanopores. The tip region is reported to have a small built-in electrostatic poten-

tial,^{23–25,59–63}and therefore is a major contributor to the observed ionic current recti-

ﬁcation behavior, suggesting that photoacids in this region may be more likely to

facilitate ionic photovoltaic action. Because the net charge in the tip of a nanopore

affects its ionic current–potential behavior, cyclic voltammetry as a function of solu-

tion pH of PET₀/1, and PET₈/1 in order to make a direct comparison to the data in

Figure 4, were used to assess the net charge in the tips of nanopores (Fig-

ure 5).^{26,58,64–66}Under acidic conditions (pH 1.5–2.0), current rectiﬁcation consistent

with a net positively charged tip of PET_n/1 was observed. This behavior strongly sug-

gests that at least some of 1 are covalently bound in the tip of the nanopore, each

with at least one free cationic ammonium group and a neutral hydroxyl group. As

the pH was titrated to more basic values (pH 2.0–5.5), the current–potential

response became ohmic and highly resistive. This behavior occurred at pH values

smaller than those measured during deprotonation of the hydroxyl group of 1 bound

Chem 5, 1648–1670, June 13, 2019 1655

Figure 6. Distribution and Binding Mode of Polymer-Bound Photoacids from pH Dependence of

Rectiﬁcation Ratio

(A) Rectiﬁcation ratio values as a function of pH for PET_nimmersed in aqueous 100 mM KCl

solutions.

(B) Representative arrangement of species in the modiﬁed tip region of PET₈/1 at pH 6.9 based on

values obtained from non-linear least-squares best ﬁt of the PET₈/1 data in (A) to Equation S1. Blue

colored numbers located above functional groups correspond to their approximate pK_avalues

obtained from inﬂection points of the best-ﬁt curve. The n values for the non-linear least-squares

best ﬁt to Equation S1 are listed in Table S2.

See also Figures S6–S9 and Tables S2 and S3.

to PET₈/1 (Figure 4), and therefore, this process is reasonably ascribed to deproto-

nation of unreacted terephthalic acid groups in the tip of the nanopore, which aided

in charge neutralization of the ammonium groups. Upon further titration of the pH

toward more alkaline conditions (pH 5.5–10) current rectiﬁcation returned but this

time the behavior was consistent with a net negatively charged tip of the nanopore,

likely caused by deprotonation of the hydroxyl group of 1. These positive–neutral–

negative surface features were consistently observed during the titration of another

eleven individual PET₀/1 (see Figure S6).

The titration data were further analyzed to determine pK_avalues and the binding

mode and binding coverage of 1 to the nanopore surface. Consistent with protocols

in previous reports, extreme current values at G4 V (see Figure S8) were converted

into rectiﬁcation ratios, normalized, and plotted as a function of pH (Figure 6A).^66–70

Data obtained from PET₀/1 lacked the sigmoidal character expected for this analysis,

likely because a single nanopore does not have a sufﬁcient ensemble of molecules

required for accurate ﬁts to the Hill equation (Equation 1). Instead, data from the

ensemble of pores in PET₈/1 were best ﬁtted to a linear combination of three Hill

equations (see Equation S1). Three Hill equations were required to account for

changes in surface charge due to deprotonation of carboxylic acid, hydroxide,

and primary ammonium moieties (Figure 6). Inﬂection points of the best-ﬁt curve

to the rectiﬁcation ratio data likely represent pK_avalues of functional groups that

line the inside tips of the nanopores. The pK_avalues were approximated to be

ꢀ3.5, ꢀ6.9, and >10 and were attributed to the carboxylic acid, hydroxyl, and

ammonium moieties, respectively. The ammonium pK_avalue could not be measured

accurately because at pH > 10 the PET polymer undergoes signiﬁcant hydrolysis.

Rectiﬁcation ratio values close to unity are consistent with conditions where there

is a nearly spatially invariant net charge of zero at the tips of the nanopores. The

pH values corresponding to this condition are termed isoelectric points and are

equal to the pH values corresponding to minimum values of the normalized rectiﬁ-

cation ratio (Figure 6A). The isoelectric points of native PET₀and PET₈were

1656 Chem 5, 1648–1670, June 13, 2019

Table 1. Acidities and Photoacidities of Photoacids

Sample

1

PET₈/1

100

[aq. KCl] (M)

Region Probed

pK_a

100

0

dissolved photoacid

5.80 G 0.03^a

nanopore tip

6.9^d

entire ﬁlm

5.1 G 0.2^a

–

nanopore tip

–

pK_a* (H^ꢂ₀)

ꢁ0.65 G 0.02^b

(ꢁ1.6 G 0.1)^b

ꢁ1.1^c

0.2^e

1.4^e

ꢁ1.4^c

Ground-state and excited-state pK_avalues for the hydroxyl group of 1 and PET_n/1. See also Figures S4

and S11 and Table S2.

^aDetermined from absorption titration data and Equation 1.

^bDetermined from photoluminescence titration data, l_ex= 379 nm, and Equations 1, 3, and 4.

^cDetermined using the Fӧrster cycle analysis and Equation 2.

^dDetermined from cyclic voltammetry titration data and Equation 1.

^eDetermined from two-photon PL microscopy, l_ex= 900 nm, and Equations 1, 3, and 4.

determined to be 3.0, which is consistent with previous reports.^58,71After modiﬁca-

tion with 1, the average isoelectric point of PET₀/1 shifted slightly more basic to 3.9,

and the isoelectric point of PET₈/1 shifted even farther to a range of 4.5–6. This large

uncertainty in the isoelectric point likely resulted from an ensemble of nanoenviron-

ments and/or distribution of pore sizes. These changes in isoelectric point strongly

suggest that the surface of the nanopore was substantially modiﬁed. Similar changes

in buffering capacities of covalently modiﬁed PET nanopores have been extensively

documented, and some exemplary modiﬁcations include amino acids^66,72and ionic

polymers.^67,68,73

Based on the best-ﬁt weightings for each Hill equation of 0.27, 0.96, and 0.27,

respectively (see Table S2), the total number of carboxylic acids and primary ammo-

niums were determined to be approximately equal and with 3.6 times more hy-

+

droxyls (3.6:1:1 ROH: R⁰NH₃R⁰⁰COOH). Because each photoacid molecule has

three amines and one hydroxyl and data support that no hydroxyls were covalently

bonded to PET, this implies that for every 10.8 amines on 1 (3.6 3 3) one amine did

not react with PET. Therefore, 91% of the amines on 1 reacted with 91% of the car-

boxylic acids on the surface of the PET nanopores ([10.8 ꢁ 1]/10.8) to covalently

bond photoacids to PET. Furthermore, because, on average, each molecule of 1

was bound to ꢀ2.7 carboxylates on PET (3 3 0.91) and native PET nanopores

have ꢀ1 COOH nm

,

^{ꢁ2 64}the binding coverage of 1 was ꢀ0.4 nm^ꢁ2(1/2.7). This value

was further substantiated by results from a two-step coupling–hydrolysis procedure

(see Figure S9; Table S3; Supplemental Experimental Procedures, Estimation of

Binding Coverage of Photoacids in Nanopores). A summary of the pK_avalues

measured in this work is presented in Table 1. Collectively, these data suggest

that the pK_avalue of the hydroxyl group of 1 is most basic when 1 is bound in the

tip of the nanopore (pK_az 6.9), followed by when 1 is dissolved in an aqueous so-

lution (pK_az 5.8) and is most acidic when 1 is tris(methylterephthalate) capped (as

photoacid 2) and dissolved in an aqueous solution (pK_az 5.4) or 1 is bound to bulk

PET polymer (pK_az 5.1). The observed ꢀ2 pH unit shift in the pK_avalue of the hy-

droxyl group of 1 when bound in the tip of a PET nanopore is consistent with an elec-

trostatic perturbation arising from a ꢀ100 mV potential drop (ꢀ2 3 2.303RT/F). This

calculated potential drop is similar to values that were previously reported in the

literature for the potential drop in the tips of PET nanopores.^25,74Moreover, this

general observation and hypothesis are consistent with those proposed to occur

in related systems containing amino acids^66,72and ionic polymers^67,68,73where, as

dictated by the Poisson equation, changes in local charge distributions alter local

Chem 5, 1648–1670, June 13, 2019 1657

electric ﬁelds, local electric potentials, local internal chemical energies, and, there-

fore, local chemical equilibria. For Brønsted acidity, this means that the value of K_a,

and therefore pK_a, differs from its value in the absence of these altered charge dis-

tributions. In addition, given the approximate net charge density of the tip of a nano-

pore modiﬁed with 1 (Figure 6B), the observed shift in pK_ato less acidic values is

consistent with a condition where less salt resides in the tip of the nanopore, as dis-

cussed in relation to the shift in pK_aof 1 dissolved in an aqueous solution containing

minimal buffer and no supporting electrolyte (Figure 1B) and analogous photoacid

molecules.

Measurement of Excited-State Photoacidities of Polymer-Bound Photoacids

Fluorescence-lifetime imaging microscopy was used to measure the excited-state

lifetime of PET₈/1 immersed in concentrated aqueous HCl and an aqueous pH 10 so-

lution. The measured excited-state lifetimes were 1.0 ns for ROH* and 1.2 ns for

RO^ꢁ*, which are shorter than the excited-state lifetimes of ROH* and RO^ꢁ* for 1 dis-

solved in an aqueous solution (3.9 ns and 5.8 ns, respectively). These differences in

lifetime suggest that when 1 is bound to PET, additional and/or faster non-radiative

relaxation pathways are likely operative.

Measurement of pK_a* in the nanopores requires that only the tip of the nanopores be

functionalized because emission from other regions of the polymers convolute the

signal. To locate 1 speciﬁcally in the tips of the nanopores, taurine (T) was used to

cap free carboxylates on the outward-facing surface of PET on the side with the small

nanopore tip opening so that they would be unreactive (PET₈/T). Similar to the solu-

tion studies, the Fӧrster cycle analysis (Equation 2) suggested that pK_a* = ꢁ1.4 in

PET₈/T+1 (see Figure S11), which is consistent with pK_a* calculated by the same

method for 1 dissolved in an aqueous solution (pK_a* = ꢁ1.1). Also analogous to 1

in solution, PET₈/T+1 that contained ROH species and immersed in weakly acidic

aqueous solutions exhibited PL from RO^ꢁ* species (Figure 7), indicating that 1 un-

derwent excited-state proton transfer and therefore was a photoacid when cova-

lently bound to the tip of a PET nanopore. However, under these weakly acidic con-

ditions, PL from ROH* was also observed for PET₈/T+1, which is in contrast to that

expected based on the PL spectra of 1 dissolved in an aqueous solution and the re-

sults from the simple Fӧrster cycle analysis. For instance, for PET₈/T+1 at pH 3, ꢀ40%

of the emission was from ROH* (Figure 7; see Figure S12) whereas for 1 in solution,

>99% of the emission was from ROH* (Figure 2). Also, results from an acid–base titra-

tion procedure with photoluminescence PL detection indicate that in contrast to the

H^ꢂ₀value of 1 dissolved in solution (H₀^ꢂ= –1.6), PET₈/T+1 exhibited pK_a* = 1.4 in the

absence of supporting electrolyte (Figure 7B; Equation 3) and pK_a* = 0.2 in the pres-

ence of 100 mM KCl supporting electrolyte (Figure 7D; Equation 3). This suggests

that surface attachment of 1 to PET and/or nanoconﬁnement alters the observed

value of pK_a*. PL spectra were completely reversible as a function of pH and isoemis-

sive points were maintained for all pH values reported. A summary of the pK_a* values

measured in this work is also included in Table 1.

To rationalize the observed excited-state behavior, several hypotheses were evalu-

ated. One hypothesis was that under focused laser excitation, the local steady-state

concentration of protons was greatly increased in comparison to the concentration

at equilibrium in the dark. To assess this hypothesis, PL from PET₈/T+1 immersed in

an aqueous pH 3 solution was measured as a function of excitation laser irradiance.

Only minor changes corresponding to a relative decrease in ROH* emission were

observed for nearly an order-of-magnitude increase in laser irradiance (see Fig-

ure S13), suggesting that this was not the cause of the observed excited-state

1658 Chem 5, 1648–1670, June 13, 2019

Figure 7. Excited-State Photoacidity of Polymer-Bound Photoacids from pH Dependence of

Photoluminescence Spectra

(A) Photoluminescence (PL) spectra of PET₈/T+1 immersed in aqueous solutions at different

Hammett acidity (H₀) or pH values and excited with a two-photon-absorption wavelength (l_ex) of

900 nm. Spectra from the excited-state species in their completely protonated (ROH*) and

deprotonated (RO^ꢁ*) forms are bolded and colored.

(B) Plot of normalized PL intensity (q) at l_maxfor each species (479 nm and 548 nm) as a function of

H₀for >1 M HCl and pH for <1 M HCl, and displayed with non-linear least-squares best ﬁts to

Equation 1, after excluding the data points shown as unﬁlled shapes.

(C) PL spectra of PET₈/T+1 immersed in aqueous 100 mM KCl solutions at different H₀or pH values

and excited with l_ex= 900 nm. Spectra from the excited-state species in their completely

protonated (ROH*) and deprotonated (RO^ꢁ*) forms are bolded and colored.

(D) Plot of normalized PL intensity (q) calculated from full spectral modeling⁵⁵as a function of H₀for >1 M HCl

and pH for <1 M HCl, and displayed with non-linear least-squares best ﬁts to Equation 1.

Correction of Observed Photoluminescence Data and Determination of Excited-State p K_aValues in

Polymers.

behavior. Another hypothesis is that differing electrostatic regions in the nanopores

resulted in a local environment with more or less of a speciﬁc type of ion, including

protons, as compared to the bulk solution. This is reasonable given that pK_a*

decreased in the presence of supporting electrolyte (Figures 7C and 7D) likely

due to additional screening of surface charges, which altered the local electrostatics

and/or the local pH of the nanopores. Moreover, both pK_aand H^ꢂ₀/pK_a* for the

hydroxyl group of 1 increased by ꢀ1 pH unit when bound to PET in the tip of the

nanopore in comparison to values measured when 1 was dissolved in solution. A ﬁnal

hypothesis is that nanoconﬁnement alters pK_avalues due to slow hydrogen bond dy-

namics that prevent water molecules from optimally solvating RO^ꢁ* and H⁺formed

Chem 5, 1648–1670, June 13, 2019 1659

from the excited-state proton-transfer step.^75,76This would result in a shift in the

excited-state quasi-equilibrium toward the protonated species and a decrease in

the yield for net proton transfer,^77–79which would be manifest as an increase in

the observed value of pK_a*. Irrespective of the mechanism, 1 is a weaker photoacid

when bound to the tip of hydrated charged PET nanopores.

Recent studies by White et al. suggest that when photoacid-modiﬁed Naﬁon is illu-

minated with visible light, ionic photovoltaic action is observed.^37,39However, the

photocurrent was at least in part limited by an increase in pK_a* for dyes bound in

the nanopores of Naﬁon versus values measured when dissolved in an aqueous so-

lution.^37,48The nanopores reported herein are nearly ideal models for these mate-

rials because the diameters of the nanopore tips are on the same sub-10-nm size

scale as hydrated domains in Naﬁon.^37,39While photocurrents from modiﬁed PET

nanopores have been previously reported,^80,81we were not able to measure reliable

photocurrents and photovoltages from our materials under any conditions, which we

expected to be the case based on the absorption spectra and photophysical prop-

erties of our materials. Nevertheless, our results above suggest that the values of pK_a

and pK_a* for 1 bound to the nanoporous tip region of PET become less acidic. This

increase in pK_a* is undesired for the ultimate application of these materials and sug-

gests that alternative photoacids with more negative values of pK_a* or ion-exchange

materials with more alkaline local pH will likely be required in order to realize an

effective combination of molecules and materials that results in large yields for

light-to-ionic energy conversion. Studies are currently underway to assess the feasi-

bility of using membranes and/or nanopores with altered surface chemistries and/or

using photoacids with more acidic pK_a* values, with larger magnitudes of DpK_a, and/

or that prefer to be localized in the center of nanochannels where the characteristics

of water more closely resemble that of bulk water.⁷⁵

Conclusions

Water-soluble photoacid molecules with terminal ammonium groups were synthe-

sized. The terminal ammonium groups allowed for facile coupling to carboxylic acids

on nanoporous PET surfaces. Spectrophotometric titration curves showed that the

hydroxyl pK_aof photoacid 1 was 5.8 when dissolved in an aqueous solution. When

1 was covalently linked to PET nanopores, electrochemical measurements and elec-

tronic absorption spectroscopy data suggest that the hydroxyl pK_aof 1 was 5.1 and

increased to 6.9 in the conﬁned tip of the nanopore. This increase in pK_ais consistent

with behavior expected when ions are excluded from the tip of the nanopore and a

larger potential affects surface-conﬁned 1. Also, a majority of 1 were attached to PET

nanopores via all three of their amine groups, the PET surface was ꢀ90% modiﬁed,

and 1 had a binding coverage of ꢀ0.4 photoacids nm^ꢁ2. Fӧrster cycle analyses and

PL titration measurements indicated that the hydroxyl pK_a* of 1 in aqueous solution

was approximately ꢁ1. Fluorescence microscopy was used to show that 1 bound to

PET was localized within nanopores and 1 still operated as a photoacid. The

apparent pK_a* was 0.2 when PET₈/1 was immersed in aqueous 100 mM KCl and

was less acidic when no supporting electrolyte was present. Moreover, titration

curves of PET_n/1 were highly non-ideal, which was postulated to be due to an aggre-

gate of conﬁnement effects, a distribution in pore geometries, and differences in

nanoenvironments. Additionally, the excited-state lifetime of

1 signiﬁcantly

decreased from about 4–5 ns to ꢀ1 ns upon binding to PET. These observations sug-

gest that less acidic environments are needed to observe excited-state proton trans-

fer in small nanopores with tip diameters < 10 nm. Use of these materials for appli-

cation in light-driven proton pumps requires further advances but could conceivably

be improved by incorporating different photoacids and/or membranes.

1660 Chem 5, 1648–1670, June 13, 2019

EXPERIMENTAL PROCEDURES

General Synthetic Procedures

Unless otherwise stated, all reactions were perfromed under an inert atmosphere of

nitrogen in oven-dried ﬂasks equipped with Teﬂon-coated magnetic stir bars. All

chemicals were obtained from commercial suppliers and used with no further puri-

ﬁcation. Reactions were monitored by time-of-ﬂight mass spectrometry or by thin-

layer chromatography using silica gel 60 F₂₅₄on aluminum sheets with ultraviolet

light for visualization.

Reagents

Pyranine (>85%, TCI America), acetic anhydride (99%, EMD Millipore), sodium

hydroxide (R95%, Macron Fine Chemicals), potassium chloride (R99%, Fisher Sci-

entiﬁc), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDC) hydrochloride (98%,

Alfa Aesar), 1-boc-ethylenediamine (98%, Combi-Blocks), thionyl chloride (R99%,

Sigma-Aldrich), propylamine (98%, Sigma-Aldrich), ethylenediamine (99%, Acros

Organics), 2-(N-morpholino)ethanesulfonic acid (MES) hydrate (99%, Oakwood

Chemical), tris(hydroxymethyl)aminomethane (R99.8%, Fisher Scientiﬁc), potas-

sium phosphate monobasic (R99.5%, Fisher Scientiﬁc), sodium tetraborate

decahydrate (R99.5%, Fisher Scientiﬁc), citric acid anhydrous (R99.5%, Fisher

Scientiﬁc), hydrochloric acid 36.5%–38% (EMD Millipore), triﬂuoroacetic acid

(99%, Oakwood Chemical), formic acid (88%, Fisher Scientiﬁc), dimethylsulfoxide

(R99.9%, EMD Millipore), taurine (99%, Sigma-Aldrich), methylterephthalate

(97%, Sigma-Aldrich), and N-hydroxysuccinimide (98% Sigma-Aldrich) were used.

Spectroscopic Characterization

Nuclear magnetic resonance (NMR) spectra were collected at 298 K at 400 MHz us-

ing a Bruker AVANCE400 spectrometer or at 500 MHz using a Bruker CRYO500

spectrometer (see Figure S14). Chemical shifts are reported using the standard d no-

tation in parts per million (ppm) relative to solvent peaks (¹H, ¹³C). Peak multiplicities

are listed as follows: s, singlet; d, doublet; t, triplet; q, quartet; and m, multiplet. The

J-coupling constants are reported in Hertz. Infrared spectra were recorded on a

Perkin-Elmer Spectrum One Fourier transform infrared (FTIR) spectrophotometer;

samples were prepared as KBr pellets. Mass spectrometric data were collected for

samples dissolved in methanol using liquid chromatography electrospray ionization

time-of-ﬂight mass spectrometry on a Waters Micromass LCT Premier mass spec-

trometer (see Figure S15).

Chemical Syntheses

Sodium 8-Acetoxypyrene-1,3,6-trisulfonate (E1)

Sodium hydroxide (0.80 g, 20 mmol) was added to a stirring solution of pyranine

(9.59 g, 18 mmol) dissolved in water (10 mL). The solution was stirred for 15 min

and then concentrated under vacuum to dryness. A heated water bath was

used to assist the distillation process. Acetic anhydride (100 mL) was added. The

suspension was heated to reﬂux and stirred for 14 h. A white powder was collected

via ﬁltration and washed with dichloromethane and acetone; the powder was

dried under vacuum (10.7 g, 18 mmol, quantitative yield). ¹H NMR (500 MHz,

D₂O) d 9.24 (s, 1H), 9.23 (d, J = 10.0, 2H), 9.17 (d, J = 10.0, 1H), 9.13 (d, J =

9.5, 1H), 8.56 (s,1H), 8.51 (d, J = 9.5, 1H), and 2.65 (s, 3H); ¹³C NMR (500 MHz,

D2O) d 173.4, 144.5, 138.3, 136.2, 136.1, 129.7, 129.0, 127.1, 126.4, 125.6,

125.5, 125.3, 125.0, 125.0, 124.7, 123.4, 120.2, and 20.5; FTIR (KBr pellet) n

3,470 (st) cm^ꢁ1, 2,959 cm^ꢁ1, 1,736 (m) cm^ꢁ1, 1,569 (m) cm^ꢁ1, 1,206 (st) cm^ꢁ1

and 1,060 (st) cm^ꢁ1

,

.

Chem 5, 1648–1670, June 13, 2019 1661

3,6,8-Tris(chlorosulfonyl)pyren-1-yl acetate (E2)

While stirring, pyrene E1 (4.352 g, 7.684 mmol) was slowly added to a solution of thi-

onyl chloride (10 mL) containing three drops of N,N-dimethylformamide. The reac-

tion was reﬂuxed for 5 h. The crude product was poured onto an aqueous ice slurry

and dissolved with dichloromethane. The organic layer was separated and concen-

trated in vacuo to obtain a yellow powder (3.312 g, 5.400 mmol, 78%). ¹H NMR

(500 MHz, CDCl₃) d 9.68 (d, J = 10.0, 1H), 9.62 (s, 1H), 9.51 (d, J = 10.0, 1H), 9.45

(d, J = 9.7, 1H), 8.91 (s, 1H), 8.83 (d, J = 9.7, 1H), and 2.67 (s, 3H); ¹³C NMR

(500 MHz, CDCl₃) d 168.7, 147.2, 140.8, 137.4, 137.2, 133.1, 131.9, 129.9, 128.2,

128.1, 127.6, 126.5, 126.1, 125.5, 125.4, 124.7, 123.4, and 21.3.

Tert-butyl-(2-((6,8-bis(N-(2-((tert-butoxycarbonyl)amino)ethyl)sulfamoyl)-3-

hydroxypyrene)-1-sulfonamido)ethyl)carbamate (E3)

N-boc-ethylenediamine (4.5704 g, 28.52 mmol) was added to a solution of H₂O

(2 mL), tetrahydrofuran (4 mL), and acetonitrile (14 mL). While stirring, pyrene E2

(2.184 g, 3.938 mmol) was added. An additional solution of H₂O (2 mL), tetrahydro-

furan (4 mL), and acetonitrile (14 mL) was added. The resulting red-violet solution

was stirred for 19 h at room temperature. Consequently, NaOH pellets (0.3 g)

were added to the mixture and stirred for 2.5 h before additional NaOH pellets

(0.4 g) were added. After 1.5 h, 100 mL of chloroform and 100 mL of 1 M NaOH

were added and vigorously shaken (with occasional venting) until the red product

dissolved completely into the aqueous layer. The aqueous solution was washed

with chloroform (200 mL 3 4). The aqueous layer was separated, ice was added

directly to the solution, and then acidiﬁed with a chilled solution of HCl (concen-

trated HCl was added directly to the ice in a 1:3 volume ratio). The resulting yellow

precipitate was collected via vacuum ﬁltration and dried under vacuum (3.224 g,

3.643 mmol, 92%). ¹H NMR (500 MHz, CD₃OD) d 9.21 (s, 1H), 9.18 (d, J = 9.8, 1H),

8.90 (d, J = 9.6, 1H), 8.94 (d, J = 9.8, 1H), 8.80 (d, J = 9.6, 1H), 8.30 (s, 1H), 2.98 (coa-

lescing m, 12H), and 1.15 (coalescing s, 27H).

2,2⁰,2⁰’-((8-Hydroxypyrene-1,3,6-trisulfonyl)tris(azanediyl))tris(ethan-

1-aminium) tris(triﬂuoroacetate) (1)

Triﬂuoroacetic acid (10 mL) was diluted with dichloromethane (50 mL) and then

chilled on ice. Pyrene E3 (3.088 g, 3.489 mmol) was then added to the chilled solu-

tion, which was then removed from the ice and stirred for 1 h. The reaction mixture

was concentrated in vacuo. Residual triﬂuoroacetic acid was co-distilled with 5 3

20 mL dichloromethane and the resulting orange powder was dried further under

high vacuum. For further puriﬁcation, the powder was dissolved in a solution of

1:19 CH₃CN:H₂O (v/v) spiked with 0.1% triﬂuoroacetic acid. The solution was passed

through a 0.2 mm ﬁlter and then injected into a preparatory reverse-phase high-per-

formance liquid chromatography instrument where the gradient was set from 5%

CH₃CN (aq, 0.1% triﬂuoroacetic acid) to 90% CH₃CN (aq, 0.1% triﬂuoroacetic

acid) over 30 min. The ﬂow rate was set to 15 mL/min. Pyrene 1 was collected

from 15% to 33% CH₃CN. Note that the preparatory reverse-phase high-perfor-

mance liquid chromatography instrument used has a maximum capacity of ꢀ0.8

grams of solute; therefore, the product was divided into 4 equal portions, each

less than 0.8 grams. Each portion was dissolved in 4.5 mL of the solution of 1:19

CH₃CN:H₂O (v/v), and a total of 4 runs were made to collect the entire batch of prod-

uct. The desired fractions were concentrated in vacuo to obtain a yellow solid (3.201

g, 3.454 mmol, 99%).¹H NMR (600 MHz, DMSO-D₆) d 9.19 (d, J = 9.7, 1H), 9.15

(s, 1H), 9.02 (d, J = 9.6, 1H), 8.92 (d, J = 9.7, 1H), 8.84 (d, J = 9.6, 1H), 8.76 (broad,

1H), 8.69 (broad, 1H), 8.66 (broad, 1H), 8.41 (s, 1H), 7.90 (broad, 9H), 3.07 (broad q,

J = 5.9, 2H), 3.01 (broad m, 4H), 2.86 and 2.84 (coalescing, 6H); ¹³C NMR (600 MHz,

1662 Chem 5, 1648–1670, June 13, 2019

DMSO-D₆) d 158.6 (q, J = 126, triﬂuoroacetate), 158.5 (q, J = 126, triﬂuoroacetate),

155.0, 136.9, 131.7, 130.5, 130.4, 130.3, 128. 6, 128.1, 126.2, 125.8, 125.5, 123.3,

121.2, 120.4, 118.4, 117.1 (q, J = 1188, triﬂuoroacetate), 117.1 (q, J = 1188, triﬂuor-

oacetate), 115.7, 40.0, 39.9, and 38.6. The m/z ratio for [M+H]⁺was calculated to be

585.1260 and was found to be 585.1270, with an acceptable D = 2.9 mDa, and an

observed D = 1.0 mDa.

Electronic Absorption Characterization of 1 in Aqueous Buffered Solution

Equal amounts of 1 were portioned into three separate volumetric ﬂasks of the same vol-

ume. Sodium tetraborate decahydrate (0.1 M), citric acid anhydrous (0.1 M),

tris(hydroxymethyl)aminomethane (0.1 M), potassium phosphate monobasic (0.1 M),

and potassium chloride (0.1 M) were added. The three ﬂasks were ﬁlled with 12 M HCl,

2 M NaOH, and distilled water (weakly acidic) to the volumetric marker. Solutions were

stirred until dissolved. The weakly acidic solution was titrated with the 12 M HCl or

2 M NaOH buffered solutions to reach the desired pH values from 2 to 12. The pH was

measured using a Fisher Scientiﬁc pH/Ion 510. The pH meter does not accurately read

pH values more acidic than pH 1 and more alkaline than pH 12. Therefore, solutions

with pH values greater than 1 and more alkaline than 12 were made from the aliquots

of the 12 M HCl and 2 M NaOH solutions, and the pH value was calculated based on

the portioning. Ultraviolet–visible electronic absorption spectra were recorded on a

Cary 60 UV-Vis Spectrophotometer. Samples consisted of aliquots dispensed in a quartz

cuvette (1 cm pathlength) and were kept at room temperature. Spectra were subtracted

byablank bufferingsolution. Thetitration wasperformedfourtimesinordertoobtainthe

mean and standard deviation of the data. The procedure was also replicated using a less

extreme buffer system that contained 1 mM potassium dihydrogen phosphate and no

supporting electrolyte (see Figure S1).

Photoluminescence Characterization of 1 in Acidic Aqueous Solution

Equal amounts of 1 were portioned into two separate volumetric ﬂasks. The two

ﬂasks were made to contain concentrated HCl and deionized water titrated with

concentrated aqueous HCl to pH 3. Ultraviolet–visible electronic absorption spectra

were recorded for both samples. Samples consisted of aliquots dispensed in a quartz

cuvette (1 cm pathlength) and were kept at room temperature. The smallest differ-

ence in absorbance occurred at the wavelength of 379 nm and was 0.009 units,

and thus 379 nm was chosen as the excitation wavelength for PL titration studies.

Further samples were prepared by mixing the two solutions. A known volume of

pH 3 solution was titrated using known volumes of concentrated aqueous HCl, which

allowed the calculation of intermediate pH values. For each sample, an ultraviolet–

visible electronic absorption spectrum and PL emission spectra were recorded. The

ultraviolet–visible electronic absorption spectra were recorded on a Cary 50 Ultravi-

olet-Visible Spectrophotometer. PL emission spectra were recorded using a Cary

Eclipse Spectrophotometer using the following excitation and emission parameters:

l_ex= 379 nm, slit_ex= 5 nm, slit_em= 5 nm, and PMT = 600 V. The spectra were re-

ported after corrections for changes in refractive indices, inner ﬁlter effects, and dif-

ferences in lifetimes of the excited-state species (see Figure S2; Supplemental

Experimental Procedures, Correction of Observed Photoluminescence Data). The

titration was performed three times in order to obtain the mean and standard devi-

ation of the data.

Preparation of PET₈/1 with Cylindrical Pores Used in Electronic Absorption

Measurements

PET ﬁlms containing 10⁸approximately cylindrical pores per cm², which were large

enough that they did not rectify ionic current, were fabricated by a procedure that is

Chem 5, 1648–1670, June 13, 2019 1663

similar to a previously described track-etching technique.^57,58,82Brieﬂy, each tem-

plate was prepared by bombarding a 12 mm thick PET ﬁlm with heavy uranium

ions whose energy of 11.1 MeV per nucleon generated a single latent track per

ion that spanned the thickness of the ﬁlm (UNILAC, GSI Helmholtz Centre for Heavy

Ion Research, Darmstadt, Germany). Cylindrical pores were then produced by an

alkaline chemical etching procedure. For better etching results, the ﬁlms were ﬁrst

irradiated with longwave ultraviolet light on both sides and were then serially rinsed

with isopropanol, methanol, and deionized water. The nonporous transparent tem-

plates were stirred in aqueous 9 M NaOH at ꢀ80^ꢃC for 15 min to create a porous

white opaque ﬁlm.

Successful coupling of 1 was achieved through a two-step procedure using condi-

tions similar to those described by Nakajima and Ikada for 1-ethyl-3-(3-dimethylami-

noproypl)carbodiimide (EDC) chemistry in aqueous media.⁸³First, the ﬁlm was

rinsed with an unbuffered aqueous solution, titrated with concentrated aqueous

HCl to pH 4. Next, the ﬁlm was placed in an aqueous solution of 0.1 M EDC,

0.1 M N-hydroxysuccinimide, and 0.1 M MES buffer (pH 5.0, titrated with aqueous

HCl) to activate the surface i.e., convert the carboxylic acids on the PET surface

into o-acylisourea esters. The ﬁlms were rinsed with distilled water and then stirred

in an aqueous solution of 0.05 M 1, pH 9 (pre-titrated with aqueous NaOH solution)

for R24 h.

Electronic Absorption Measurements on PET₈/1 with Cylindrical Pores

An aqueous solution of 0.1 M sodium tetraborate decahydrate, 0.1 M citric acid

anhydrous, 0.1 M tris(hydroxymethyl)aminomethane, 0.1 M potassium phosphate

monobasic, and 0.1 M potassium chloride was produced. The solution was titrated

with concentrated aqueous HCl or 9 M NaOH to the desired pH, as measured using a

Fisher Scientiﬁc pH/Ion 510. After titrating the solution to the desired pH, the ﬁlm

was stirred for ꢀ1 min in the buffered solution. The ﬁlm was removed and while drip-

ping wet, pushed ﬂush against the aperture. Ultraviolet–visible electronic absorp-

tion spectra were recorded on a Cary 60 UV-Vis Spectrophotometer. To account

for the inhomogeneity of the ﬁlm, each reported spectrum is an average of 10 scans

taken at different locations on the ﬁlm. Lastly, the spectra were baselined with PET₈

modiﬁed with ethylenediamine.

Preparation of Single Conical Nanopores Used in Electrochemical

Measurements

Conical pores in PET ﬁlms were fabricated by a previously described track-etching

technique.^57,58,82Brieﬂy, each template was prepared by bombarding a 12 mm thick

PET ﬁlm with heavy uranium ions whose energy of 11.4 MeV per nucleon for PET₀or

11.1 MeV per nucleon for PET₈generated a single latent track per ion that spanned

the thickness of the ﬁlm (UNILAC, GSI Helmholtz Centre for Heavy Ion Research,

Darmstadt, Germany). Conical nanopore(s) were then produced by an alkaline

chemical etching procedure. For better etching results, the ﬁlms were ﬁrst irradiated

with longwave ultraviolet light on both sides and were then serially rinsed with iso-

propanol, methanol, and deionized water. The ﬁlm was then placed in a clean

two-chamber cell for subsequent etching: aqueous 9 M NaOH was placed on one

side of the ﬁlm to etch the latent track, while an aqueous stopping medium contain-

ing 1 M formic acid and 1 M KCl was placed on the other side of the ﬁlm to halt the

reaction. A platinum (Pt) electrode was placed on each side of the membrane and a

1 V bias was applied to the anode, which was immersed in the aqueous 9 M NaOH

solution. An increase in current, indicating a decrease in the resistance across the

ﬁlm, signiﬁed pore breakthrough and completed synthesis of a conical nanopore.

1664 Chem 5, 1648–1670, June 13, 2019

The reaction for PET₀was usually halted when a current of ꢀ200 pA was observed.

The resulting nanopore had a large base opening with diameter A, which was

approximated based on the bulk etch rate for the described etching scenario of

2.13 nm/min. The diameter of the small tip opening, a, was approximated by the

following equation:

4L

R =

;

(Equation 5)

pgAa

where R is the resistance of a nanopore, L is the length of the pore, and g is the spe-

ciﬁc conductivity of the electrolyte, which was 10 S/m for an aqueous 1 M KCl solu-

tion. R was determined by calculating the slope of the linear current–potential data

between ꢁ0.1 V and +0.1 V to obtain the electrical conductance and then recipro-

cating it.

A PET₀ﬁlm was then placed in a two-chamber cell that was subsequently rinsed with

a pH 4.0, 100 mM KCl aqueous electrolyte. An aqueous solution of 0.1 M EDC, 0.1 M

N-hydroxysuccinimide, and 0.01 M MES buffer was titrated to pH 5.0 with dilute

aqueous HCl. The EDC solution was placed on both sides of the ﬁlm and allowed

to react for 1 h. Without disassembling the cell, the EDC solution was then removed

and electrochemical cell chambers were rinsed twice with deionized water. In a sepa-

rate beaker, an aqueous 0.05 M 1 solution was titrated to pH 9.0 and placed on both

sides of the ﬁlm for R24 h. Without disassembling the cell, the solution was removed

and the ﬁlm was rinsed ﬁve times with deionized water. The membrane was stored in

the cell using an aqueous pH 7.0 electrolyte containing 100 mM KCl and 1 mM

tris(hydroxymethyl)aminomethane.

Electrochemical Characterization on PET₀/1 and PET₈/1

Experiments were carried out using a VSP-300 potentiostat (Bio-Logic Science)

equipped with low-current cables and a Faraday cage (Bio-Logic Science). The

PET_n/1 was placed in a two-chamber cell and a four-electrode setup was utilized.

Two Pt wires served as the working and counter electrodes, while two standard

calomel electrodes (SCEs) (KCl saturated) served as the reference electrodes. One

of each was placed in chambers on either side of the membrane.

An aqueous 0.1 M KCl stock solution was purged via stirring and the vigorous

bubbling of argon. The pH was continuously measured via a Fisher Scientiﬁc pH/

Ion 510 probe that was immersed in the solution. The stock solution was purged

for 20 min before performing the initial cyclic voltammetric titration experiment.

The assembled two-chamber cell was rinsed with deionized water ﬁve times and

then carefully rinsed with purged electrolyte solution ﬁve times. Cyclic voltammo-

gram were recorded using the following parameters: a scan rate of 50 mV/s, scan

range of 4 V to ꢁ4 V versus reference, the scan started and ended at 0 V versus

reference, the scan cycle was repeated three times, current was measured over

the last 50% of the step duration, recorded current was averaged over 10 voltage

steps, the voltage resolution was set to 200 mV, and the current resolution range

was set to 10 nA. The electrolyte solution in the cell was not purged during the

measurement to prevent interference from bubbles. However, the stock solution

was continuously purged during the cyclic voltammogram measurements. Titra-

tions employed aqueous solutions of either HCl or KOH. Salt buildup was attenu-

ated by titrating one purged stock solution from pH ꢀ6 to ꢀ1 with only HCl and

titrating another purged stock solution from pH ꢀ6 to ꢀ11 with only KOH. Mea-

surements were performed from pH ꢀ6 to ꢀ1 and then from pH ꢀ6 to ꢀ11.

Chem 5, 1648–1670, June 13, 2019 1665

Determining Binding Coverage of Photoacid in Nanopores

As described above, conical pores in a PET₀ﬁlm and PET₈ﬁlm were prepared, and

the two ﬁlms were completely modiﬁed with 1 to form PET₀/1 and PET₈/1. Sepa-

rately, 1 was hydrolyzed from both modiﬁed ﬁlms using 5 mL of aqueous 1 M

NaOH. Electronic absorption spectra of both samples were recorded on a Cary

60 UV-Vis Spectrophotometer. The binding surface coverage of 1 inside a single

nanopore was calculated using Equation S5.

Synthesis and Electronic Absorption Measurements on Trimethyl

4,4⁰,4⁰⁰-(((((8-Hydroxypyrene-1,3,6-trisulfonyl)tris(azanediyl))tris(ethane-2,1-

diyl))tris(azanediyl))tris(carbonyl))tribenzoate (2)

Methylterephthalate (0.006 g, 3 mmol), EDC (0.008 g, 4 mmol), N-hydroxysuccini-

mide (0.006 g, 5 mmol), and MES buffer (0.005 g, 3 mmol) were added to a mixture

of 5 mL of water and 5 mL of methanol. The reaction was stirred at room temperature

for 1.5 h. Then, 1 (0.006 g, 0.6 mmol) was added and stirred at room temperature for

0.5 h. The reaction vessel was heated to 60^ꢃC and stirred for 4 days. The solution was

diluted with deionized water so that major absorption peaks were ꢀ0.3 units. Elec-

tronic absorption spectra were recorded on a Cary 60 UV-Vis Spectrophotometer

and baselined with deionized water. The cocktail containing 2 was titrated using

concentrated aqueous HCl and 9 M NaOH to minimize changes to the concentration

of 2.

Preparation of PET₈/1 with Conical Pores Used in Fluorescence Microscopy

Measurements

PET₈was prepared by the track-etching technique described earlier with the excep-

tion that the etchant process was usually halted when a current of ꢀ1 mA was

observed instead of ꢀ200 pA. Even though PET₈contained a factor of ꢀ10⁸times

more heavy-ion irradiated regions, stopping the etchant process at this small of a

current value ensured optimal rectiﬁcation behavior.

The outward-facing surface of PET₈on the side with the small nanopore tip open-

ing—not in the conical nanopores—was modiﬁed with non-ﬂuorescent taurine (T)

molecules followed by 1 such that 1 was located primarily inside the tips of the

nanopores (PET₈/T+1). PET₈was placed in a two-chamber cell and the cell was

rinsed with an aqueous pH 4.0 electrolyte containing 100 mM KCl. Deionized water

was placed on the large-pore side of the ﬁlm. An aqueous solution of 0.1 M EDC,

0.1 M N-hydroxysuccinimide, and 0.01 M MES buffer was titrated to pH 5.0 with

dilute aqueous HCl. The EDC solution was placed at the tip side of the ﬁlm and al-

lowed to react for 0.5 min. Without disassembling the cell, the EDC solution was

then removed and electrochemical cell chambers were rinsed twice with deionized

water. In a separate beaker, an aqueous 0.05 M T solution was titrated to pH 9.0

and placed on both sides of the ﬁlm for R4 h. The cell was rinsed with a pH 4.0,

100 mM KCl aqueous electrolyte. Deionized water was placed on the base side

of the ﬁlm. An aqueous solution of 0.1 M EDC, 0.1 M N-hydroxysuccinimide, and

0.01 M MES buffer was titrated to pH 5.0 with dilute aqueous HCl. The EDC solu-

tion was placed at the small tip side of the ﬁlm and allowed to react for 20 min.

Without disassembling the cell, the EDC solution was then removed and electro-

chemical cell chambers were rinsed twice with deionized water. In a separate

beaker, an aqueous 0.05 M 1 solution was titrated to pH 9.0 and placed on both

sides of the ﬁlm for R24 h. Without disassembling the cell, the solution was

removed, and the ﬁlm was rinsed with deionized water ﬁve times. The membrane

was stored in the cell and wetted with an aqueous pH 4.0 electrolyte containing

100 mM KCl.

1666 Chem 5, 1648–1670, June 13, 2019

Two-Photon Fluorescence Microscope Setup for Spectral Imaging

A custom-built two-photon-excitation microscope based on an Olympus FLUOVIEW

FV1000 (Olympus Corporation, Tokyo, Japan) was used for all ﬂuorescence micro-

scopy experiments. The microscope was equipped with an Olympus UPlanApo

603 water-immersion objective (numerical aperture = 1.2). A mode-locked 80

MHz Ti:Sapphire Chameleon Ultra laser (Coherent, Santa Clara, CA) tunable in the

range from 690 nm to 1,040 nm was used for multiphoton excitation. The laser ﬂu-

ence was controlled by an acoustic optical modulator driven by SimFCS (LFD,

UCI). A grating-based spectrograph (Andor Shamrock SR-303i) with a 512 channel

ultrafast EMCCD (Andor iXon Ultra) was used to collect a spectrum from every pixel

with a grating dispersion of 50 lines per meter and a blaze wavelength of 600 nm. In

order to ensure spectral acquisition was synced to the frame clock and sampled to

the pixel clock, the Olympus FV1000 galvanometer scanning mirrors were driven

externally by SimFCS (LFD, UCI) using a data acquisition controller board (IOTECH

DAQboard 3001). The spectral images of 2563256 pixels were collected with a pixel

dwell time of 32 microseconds. The spectra are corrected with a spectral response

curve of the system calibrated using a tungsten lamp, by a technique reported by

Chen et al.⁸⁴The multiphoton imaging measurements were carried out with an

open pinhole, and the luminescent photons from the microscope were delivered

to the spectrograph using a 200 mm multimode optical ﬁber.

Data Processing of Spectral Images from Two-Photon Fluorescence

Microscopy

All data were obtained and analyzed using SimFCS software (Laboratory for Fluores-

cence Dynamics, UCI, http://www.lfd.uci.edu/globals/, Irvine, CA).

Sample Preparation for Two-Photon Fluorescence Microscopy

PET ﬁlms were cut to <0.5 cm 3 <0.5 cm rectangles and placed ﬂat on a glass bot-

tom microwell dish (MatTek Corporation) with the small tip side of the nanopores

face down. A titrated solution was pipetted on the ﬁlm, which did not contain buffer

or salt, but instead was pre-titrated with concentrated aqueous solutions of either

HCl or NaOH. The ﬁlm was completely immersed and the solution was pipetted

back and forth over the ﬁlm several times as a means to mix the solution into the

nanopores. A glass microscope coverslip was cut and pressed on the top of the

ﬁlm. The solution that protruded from the coverslip side was pipetted away. Enough

solution was present in between the glass microwell dish and glass coverslip to

completely wet the nanopores.

Unmodiﬁed PET and PET₈whose outward-facing surface on the side with the small

nanopore tip opening was modiﬁed with T molecules via carbodiimide chemistry

(PET₈/T) displayed excitation-wavelength-dependent emission (see Table S4). An

excitation wavelength of 900 nm was used to excite the single-photon isosbestic

point (ꢀ450 nm, Figure 1A). Raw emission spectra from PET₈/T+1 were individually

subtracted by scaled emission spectra of PET₈/T.

SUPPLEMENTAL INFORMATION

Supplemental Information can be found online at https://doi.org/10.1016/j.chempr.

2019.04.022.

ACKNOWLEDGMENTS

The authors are grateful for ﬁnancial support from the Department of Chemistry and

the School of Physical Sciences at the University of California Irvine, Research

Chem 5, 1648–1670, June 13, 2019 1667

Corporation for Science Advancement under a Cottrell Scholar Award (award

#24169) and the Gordon and Betty Moore Foundation under a Moore Inventor

Fellowship (GBMF grant #5641) for S.A. The membranes employed in this work

were irradiated during the experiment UMAT, at the beam line X0, at the GSI Helm-

holtzzentrum fuer Schwerionenforschung, Darmstadt, Germany in the frame of FAIR

Phase-0. We thank the UCI Laser Spectroscopy Facility (LSF) for use of photolumines-

cence spectroscopy instrumentation, the NMR Facility for NMR measurements, and

the Mass Spectrometry Facility for mass spectrometry measurements. We also thank

Maria Eugenia Toimil-Molares for supplying the heavy-ion irradiated PET ﬁlms and

discussing sample preparation, Zuzanna Siwy and Steven Buchsbaum for training

and guidance in nanopore etching and functionalization, Enrico Gratton for use of

the Laboratory of Fluorescence Dynamics (LFD) grant number NIH P41-

GM103540, Jonathan Lam for general synthetic advice, and Alan Heyduk for use

of his laboratory space and his group members for support and guidance.

AUTHOR CONTRIBUTIONS

S.A. proposed the research. C.D.S. synthesized molecules, prepared samples, per-

formed measurements, and analyzed the data with advice from S.A., J.V.C., and

M.D. S.A. and C.D.S. prepared the manuscript.

DECLARATION OF INTERESTS

S.A. and C.D.S. are co-inventors on the following pending patent applications that

are relevant to this manuscript: US20180065095A1 and WO2018049061A1. S.A. re-

ceives sponsored research funding support by Nissan Chemical Corporation for

work related to that reported in this manuscript.

Received: September 17, 2018

Revised: January 8, 2019

Accepted: April 24, 2019

Published: June 6, 2019

REFERENCES AND NOTES

1. Memming, R. (2015). Semiconductor

Electrochemistry, Second Edition (Wiley-VCH).

8. Glaaser, I.W., and Slesinger, P.A. (2017). Dual

activation of neuronal G protein-gated

inwardly rectifying potassium (GIRK) channels

by cholesterol and alcohol. Sci. Rep 7 , 4592.

Comparison of bipolar membranes by means

of chronopotentiometry. J. Memb. Sci. 199 ,

177–190.

2. Yasui, M., Hazama, A., Kwon, T.H., Nielsen, S.,

Guggino, W.B., and Agre, P. (1999). Rapid

gating and anion permeability of an

15. Chen, P., Mitsui, T., Farmer, D.B., Golovchenko,

J., Gordon, R.G., and Branton, D. (2004).

Atomic layer deposition to ﬁne-tune the

surface properties and diameters of fabricated

nanopores. Nano Lett. 4 , 1333–1337.

9. Lee, I., Lee, J.W., and Greenbaum, E. (1997).

Biomolecular electronics: vectorial arrays of

photosynthetic reaction centers. Phys. Rev.

Lett. 79 , 3294–3297.

intracellular aquaporin. Nature 402 , 184–187.

3. Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene,

M., Chait, B.T., and MacKinnon, R. (2003). X-ray

structure of a voltage-dependent K⁺channel.

Nature 423 , 33–41.

10. Simons, R., and Khanarian, G. (1976). Water

dissociation in bipolar membranes -

experiment and theory. J. Membr. Biol. 38 ,

11–30.

16. Li, J., Stein, D., McMullan, C., Branton, D., Aziz,

M.J., and Golovchenko, J.A. (2001). Ion-beam

sculpting at nanometre length scales. Nature

412 , 166–169.

4. Bean, B.P. (2007). The action potential in

mammalian central neurons. Nat. Rev.

Neurosci 8 , 451–465.

11. Bassignana, I.C., and Reiss, H. (1983). Ion

transport and water dissociation in bipolar ion

exchange membranes. J. Memb. Sci. 15 , 27–41.

17. Kong, Y., Fan, X., Zhang, M., Hou, X., Liu, Z.,

Zhai, J., and Jiang, L. (2013). Nanoﬂuidic diode

based on branched alumina nanochannels with

tunable ionic rectiﬁcation. ACS Appl. Mater.

Interfaces 5 , 7931–7936.

5. And e ´ rasson, L.E., and Vanng a ˚r d, T. (1998).

¨

Photosystems I and II. Annu. Rev. Plant Physiol.

Plant Mol. Biol. 39 , 379–411.

12. Ramırez, P., Rapp, H.-J., Maf e ´ , S., and Bauer, B.

´

(1994). Bipolar membranes under forward and

reverse bias conditions. Theory vs. experiment.

J. Electroanal. Chem. 375 , 101–108.

6. Furshpan, E.J., and Potter, D.D. (1959).

Transmission at the giant motor synapses of

the crayﬁsh. J. Physiol 145 , 289–325.

18. Siwy, Z., Dobrev, D., Neumann, R., Trautmann,

C., and Voss, K. (2003). Electro-responsive

asymmetric nanopores in polyimide with

stable ion-current signal. Appl. Phys. A 76 ,

781–785.

13. Maf e ´ , S., and Ram ı ´rez, P. (1997).

Electrochemical characterization of polymer

ion-exchange bipolar membranes. Acta Polym

48 , 234–250.

7. Phelan, P., Goulding, L.A., Tam, J.L.Y., Allen,

M.J., Dawber, R.J., Davies, J.A., and Bacon, J.P.

(2008). Molecular mechanism of rectiﬁcation

at identiﬁed electrical synapses in the

Drosophila giant ﬁber system. Curr. Biol. 18 ,

1955–1960.

19. Siwy, Z., Apel, P., Baur, D., Dobrev, D.D.,

Korchev, Y.E., Neumann, R., Spohr, R.,

Trautmann, C., and Voss, K.O. (2003).

14. Wilhelm, F.G., van der Vegt, N.F.A.,

Strathmann, H., and Wessling, M. (2002).

1668 Chem 5, 1648–1670, June 13, 2019

Preparation of synthetic nanopores with

transport properties analogous to biological

channels. Surf. Sci. 532–535 , 1061–1066.

with p -stacked ﬂuorophore scaffolds in lipid

bilayers. Science 313 , 84–86.

for ultrasensitive ﬂuorescence spectroscopy.

Photochem. Photobiol. Sci. 13 , 548–562.

51. Das, I., Panja, S., and Halder, M. (2016).

Modulation and salt-induced reverse

modulation of the excited-state proton-

transfer process of lysozymized pyranine: the

contrasting scenario of the ground-state acid-

base equilibrium of the photoacid. J. Phys.

Chem. B 120 , 7076–7087.

36. Xie, X., Crespo, G.A., Mistlberger, G., and

Bakker, E. (2014). Photocurrent generation

based on a light-driven proton pump in an

artiﬁcial liquid membrane. Nat. Chem. 6 ,

202–207.

20. Chtanko, N., Molares, M.E.T., Cornelius, T.,

Dobrev, D., and Neumann, R. (2004). Etched

single-ion-track templates for single nanowire

synthesis. J. Phys. Chem. B 108 , 9950–9954.

21. Kalman, E., Healy, K., and Siwy, Z.S. (2007).

Tuning ion current rectiﬁcation in asymmetric

nanopores by signal mixing. Europhys. Lett. 78 ,

28002.

37. White, W., Sanborn, C.D., Reiter, R.S., Fabian,

D.M., and Ardo, S. (2017). Observation of

photovoltaic action from photoacid-modiﬁed

naﬁon due to light-driven ion transport. J. Am.

Chem. Soc. 139 , 11726–11733.

52. Hercules, D.M., and Rogers, L.B. (1959).

Luminescence spectra of naphthols

and naphthalenediols low

22. Zhang, B., Zhang, Y., and White, H.S. (2004).

The nanopore electrode. Anal. Chem. 76 ,

6229–6238.

temperature phenomena. J. Phys. Chem.

64 , 397–400.

38. Soroka, H.P., Simkovitch, R., Kosloff, A.,

Shomer, S., Pevzner, A., Tzang, O., Tirosh, R.,

Patolsky, F., and Huppert, D. (2013). Excited-

state proton transfer and proton diffusion near

hydrophilic surfaces. J. Phys. Chem. C 117 ,

25786–25797.

53. Wehry, E.L., and Rogers, L.B. (1965). Variation

of excited-state pKa values with method of

measurement. Spectrochim. Acta 21 , 1976–

1978.

23. Siwy, Z., Heins, E., Harrell, C.C., Kohli, P., and

Martin, C.R. (2004). Conical-nanotube ion-

current rectiﬁers: the role of surface charge.

J. Am. Chem. Soc. 126 , 10850–10851.

54. Ireland, J.F., and Wyatt, P.A.H. (1976). Acid-

base properties of electronically excited states

of organic molecules. Adv. Phys. Org. Chem.

12 , 131–221.

39. White, W., Sanborn, C.D., Fabian, D.M., and

Ardo, S. (2018). Conversion of visible light into

ionic power using photoacid-dye-sensitized

bipolar ion-exchange membranes. Joule 2 ,

94–109.

24. Siwy, Z., Gu, Y., Spohr, H.A., Baur, D., Wolf-

Reber, A., Spohr, R., Apel, P., and Korchev, Y.E.

(2002). Rectiﬁcation and voltage gating of ion

currents in a nanofabricated pore. Europhys.

Lett. 60 , 349–355.

55. O’Donnell, R.M., Sampaio, R.N., Li, G.,

Johansson, P.G., Ward, C.L., and Meyer, G.J.

(2016). Photoacidic and photobasic behavior of

transition metal compounds with carboxylic

acid Group(s). J. Am. Chem. Soc. 138 , 3891–

3903.

40. Wen, L., Hou, X., Tian, Y., Nie, F.Q., Song, Y.,

Zhai, J., and Jiang, L. (2010). Bioinspired

smart gating of nanochannels toward

photoelectric-conversion systems. Adv. Mater

22 , 1021–1024.

25. Siwy, Z.S. (2006). Ion-current rectiﬁcation in

nanopores and nanotubes with broken

symmetry. Adv. Funct. Mater 16 , 735–746.

26. Nguyen, G., Vlassiouk, I., and Siwy, Z.S. (2010).

Comparison of bipolar and unipolar ionic

diodes. Nanotechnology 21 , 265301.

56. Apel, P. (2001). Track etching technique in

membrane technology. Radiat. Meas. 34 ,

559–566.

41. Xiao, K., Xie, G., Li, P., Liu, Q., Hou, G., Zhang,

Z., Ma, J., Tian, Y., Wen, L., and Jiang, L. (2014).

A biomimetic multi-stimuli-response ionic gate

using a hydroxypyrene derivation-

27. Wei, C., Bard, A.J., and Feldberg, S.W. (1997).

Current rectiﬁcation at quartz nanopipet

electrodes. Anal. Chem. 69 , 4627–4633.

57. Apel, P.Y., Korchev, Y.E., Siwy, Z., Spohr, R., and

Yoshida, M. (2001). Diode-like single-ion track

membrane prepared by electro-stopping.

Nucl. Instrum. Methods Phys. Res. B 184 ,

337–346.

functionalized asymmetric single nanochannel.

Adv. Mater 26 , 6560–6565.

28. Bruckbauer, A., Ying, L., Rothery, A.M., Zhou,

D., Shevchuk, A.I., Abell, C., Korchev, Y.E., and

Klenerman, D. (2002). Writing with DNA and

protein using a nanopipet for controlled

42. Tolbert, L.M., and Solntsev, K.M. (2002).

Excited-state proton transfer: from

constrained systems to ‘‘super’’ photoacids to

superfast proton transfer. Acc. Chem. Res. 35 ,

19–27.

58. Vlassiouk, I., and Siwy, Z.S. (2007). Nanoﬂuidic

delivery. J. Am. Chem. Soc. 124 , 8810–8811.

diode. Nano Lett. 7 , 552–556.

29. Piper, J.D., Clarke, R.W., Korchev, Y.E., Ying, L.,

and Klenerman, D. (2006). A renewable

nanosensor based on a glass nanopipette.

J. Am. Chem. Soc. 128 , 16462–16463.

ꢀ

59. Siwy, Z., and Fulinski, A. (2004). A nanodevice

43. Wurfel, P. (2005). Physics of Solar Cells, First

¨

for rectiﬁcation and pumping ions. Am. J. Phys.

72 , 567–574.

Edition (Wiley-VCH).

44. Gutman, M., and Huppert, D. (1979). Rapid pH

and deltamuH+ jump by short laser pulse.

J. Biochem. Biophys. Methods 1 , 9–19.

60. Constantin, D., and Siwy, Z.S. (2007).

Poisson-Nernst-Planck model of ion current

rectiﬁcation through a nanoﬂuidic diode. Phys.

Rev. E Stat. Nonlin. Soft Matter Phys. 76 ,

041202.

30. Umehara, S., Pourmand, N., Webb, C.D., Davis,

R.W., Yasuda, K., and Karhanek, M. (2006).

Current rectiﬁcation with poly-L-lysine-coated

quartz nanopipettes. Nano Lett. 6 , 2486–2492.

45. Gutman, M., Huppert, D., and Pines, E. (1981).

The pH jump: a rapid modulation of pH of

aqueous solutions by a laser pulse. J. Am.

Chem. Soc. 103 , 3709–3713.

31. McKelvey, K., German, S.R., Zhang, Y., White,

H.S., and Edwards, M.A. (2017). Nanopipettes

as a tool for single nanoparticle

61. Xue, J.M., Zou, X.Q., Xie, Y.B., and Wang, Y.G.

(2009). Molecular dynamics simulations on the

ionic current through charged nanopores.

J. Phys. D: Appl. Phys. 42 , 105308.

electrochemistry. Curr. Opin. Electrochem 6 ,

4–9.

46. Mauritz, K.A., and Moore, R.B. (2004). State of

understanding of naﬁon. Chem. Rev. 104 ,

4535–4585.

62. Kovarik, M.L., Zhou, K., and Jacobson, S.C.

(2009). Effect of conical nanopore diameter on

ion current rectiﬁcation. J. Phys. Chem. B 113 ,

15960–15966.

32. Guerrette, J.P., and Zhang, B. (2010). Scan-rate-

dependent current rectiﬁcation of cone-

shaped silica nanopores in quartz

nanopipettes. J. Am. Chem. Soc. 132 , 17088–

17091.

47. Gaieck, W., and Ardo, S. (2014). Challenges

and opportunities for ion-exchange

membranes in solar fuels devices. Rev. Adv. Sci.

Eng 3 , 277–287.

63. Siwy, Z.S., and Howorka, S. (2010). Engineered

voltage-responsive nanopores. Chem. Soc.

Rev. 39 , 1115–1132.

33. Sun, K., and Mauzerall, D. (1996). A simple light-

driven transmembrane proton pump. Proc.

Natl. Acad. Sci. USA 93 , 10758–10762.

48. Kusoglu, A., and Weber, A.Z. (2017). New

insights into perﬂuorinated sulfonic-acid

ionomers. Chem. Rev. 117 , 987–1104.

64. He, Y., Gillespie, D., Boda, D., Vlassiouk, I.,

Eisenberg, R.S., and Siwy, Z.S. (2009). Tuning

transport properties of nanoﬂuidic devices with

local charge inversion. J. Am. Chem. Soc. 131 ,

5194–5202.

34. Steinberg-Yfrach, G., Liddell, P.A., Hung, S.-C.,

Moore, A.L., Gust, D., and Moore, T.A. (1997).

Conversion of light energy to proton potential

in liposomes by artiﬁcial photosynthetic

49. Onufriev, A., Case, D.A., and Ullmann, G.M.

(2001). A novel view of pH titration

in biomolecules. Biochemistry 40 , 3413–

3419.

reaction centres. Nature 385 , 239–241.

65. Ali, M., Schiedt, B., Healy, K., Neumann, R., and

Ensinger, W. (2008). Modifying the surface

charge of single track-etched conical

nanopores in polyimide. Nanotechnology 19 ,

085713.

35. Bhosale, S., Sisson, A.L., Talukdar, P.,

50. Finkler, B., Spies, C., Vester, M., Walte, F.,

Omlor, K., Riemann, I., Zimmer, M., Stracke, F.,

Gerhards, M., and Jung, G. (2014).

Furstenberg, A., Banerji, N., Vauthey, E., Bollot,

¨

G., Mareda, J., Roger, C., Wurthner, F., et al.

¨

(2006). Photoproduction of proton gradients

Highly photostable ‘‘super’’-photoacids

Chem 5, 1648–1670, June 13, 2019 1669

66. Ali, M., Ramirez, P., Maf e ´ , S., Neumann, R., and

Ensinger, W. (2009). A pH-tunable nanoﬂuidic

diode with a broad range of rectifying

72. Yameen, B., Ali, M., Neumann, R., Ensinger, W.,

Knoll, W., and Azzaroni, O. (2009).

Single conical nanopores displaying

pH-tunable rectifying characteristics.

Manipulating ionic transport with zwitterionic

polymer brushes. J. Am. Chem. Soc. 131 , 2070–

2071.

78. Tan, H.S., Piletic, I.R., and Fayer, M.D. (2005).

Orientational dynamics of water

conﬁned on a nanometer length

scale in reverse micelles. J. Chem. Phys. 122 ,

174501.

properties. ACS Nano 3 , 603–608.

67. Yameen, B., Ali, M., Neumann, R., Ensinger, W.,

Knoll, W., and Azzaroni, O. (2009). Synthetic

proton-gated ion channels via single

solid-state nanochannels modiﬁed with

responsive polymer brushes. Nano Lett. 9 ,

2788–2793.

79. Tan, H.S., Piletic, I.R., Riter, R.E., Levinger, N.E.,

and Fayer, M.D. (2005). Dynamics of water

conﬁned on a nanometer length scale in

reverse micelles: ultrafast infrared

73. Ali, M., Yameen, B., Cervera, J., Ram ı ´rez, P.,

Neumann, R., Ensinger, W., Knoll, W., and

Azzaroni, O. (2010). Layer-by-layer assembly

of polyelectrolytes into ionic current

rectifying solid-state nanopores: insights from

theory and experiment. J. Am. Chem. Soc. 132 ,

8338–8348.

vibrational echo spectroscopy. Phys. Rev. Lett.

94 , 057405.

68. Yameen, B., Ali, M., Neumann, R., Ensinger, W.,

Knoll, W., and Azzaroni, O. (2010).

Proton-regulated rectiﬁed ionic transport

through solid-state conical nanopores

modiﬁed with phosphate-bearing polymer

brushes. Chem. Commun. (Camb.) 46 , 1908–

1910.

80. Zhang, Z., Kong, X.Y., Xie, G., Li, P., Xiao, K.,

Wen, L., and Jiang, L. (2016). ‘‘Uphill’’ cation

transport: a bioinspired photodriven ion pump.

Sci. Adv 2 , e1600689.

74. Cervera, J., Schiedt, B., Neumann, R., Maf e ´ , S.,

and Ram ı ´rez, P. (2006). Ionic conduction,

rectiﬁcation, and selectivity in

81. Zhang, Z., Wen, L., and Jiang, L. (2018).

Bioinspired smart asymmetric

nanochannel membranes. Chem. Soc. Rev.

47 , 322–356.

single conical nanopores. J. Chem. Phys. 124 ,

104706.

69. Gilles, F.M., Tagliazucchi, M., Azzaroni, O., and

Szleifer, I. (2016). Ionic conductance

of polyelectrolyte-modiﬁed

75. Spry, D.B., Goun, A., Glusac, K., Moilanen, D.E.,

and Fayer, M.D. (2007). Proton transport and

the water environment in naﬁon fuel cell

membranes and AOT reverse micelles. J. Am.

Chem. Soc. 129 , 8122–8130.

nanochannels: nanoconﬁnement effects on

the coupled protonation equilibria of

polyprotic brushes. J. Phys. Chem. C 120 ,

4789–4798.

82. Spohr, R. (1983). Methods for producing

nuclear traces or microholes originating from

nuclear traces of an individual ion, United

States Patent 4369370.

70. Tagliazucchi, M., Azzaroni, O., and Szleifer, I.

(2010). Responsive polymers end-tethered in

solid-state nanochannels: when

76. Spry, D.B., and Fayer, M.D. (2009).

Proton transfer and proton

83. Nakajima, N., and Ikada, Y. (1995). Mechanism

of amide formation by carbodiimide for

bioconjugation in aqueous media. Bioconjug.

Chem. 6 , 123–130.

concentrations in protonated naﬁon fuel

cell membranes. J. Phys. Chem. B 113 , 10210–

10221.

nanoconﬁnement really matters. J. Am. Chem.

Soc. 132 , 12404–12411.

71. Siwy, Z., Apel, P., Dobrev, D., Neumann, R.,

Spohr, R., Trautmann, C., and Voss, K. (2003).

Ion transport through asymmetric nanopores

prepared by ion track etching. Nucl. Instrum.

Methods Phys. Res. B 208 , 143–148.

77. Piletic, I.R., Tan, H.S., and Fayer, M.D. (2005).

Dynamics of nanoscopic water: vibrational

echo and infrared pump-probe studies of

reverse micelles. J. Phys. Chem. B 109 , 21273–

21284.

84. Chen, H., Gratton, E., and Digman, M.A. (2015).

Spectral properties and dynamics of gold

nanorods revealed by EMCCD-based spectral

phasor method. Microsc. Res. Tech 78 ,

283–293.

1670 Chem 5, 1648–1670, June 13, 2019

Article Doi

Interfacial and Nanoconfinement Effects Decrease the Excited-State Acidity of Polymer-Bound Photoacids

DOI: 10.1016/j.chempr.2019.04.022

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.chempr.2019.04.022

Full text of DOI:10.1016/j.chempr.2019.04.022

Products guided by the article

R&D Labs maybe for 264270-90-8

Relevant to this article

Hot Product