Solid State Sensing of Nonpolar VOCs Using the Bistable Expansion

Jun 22, 2017 - A molecular switch that operates from the discrete expansion and contraction of helical polymer chain segments in response to chemical ...
1 downloads 7 Views 3MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Solid State Sensing of Nonpolar VOCs Using the Bistable Expansion and Contraction of Helical Polycarbodiimides Raymond Campos,†,‡ James F. Reuther,†,§ Nimmy R. Mammoottil,† and Bruce M. Novak*,† †

Department of Chemistry and Alan G. MacDiarmid NanoTech Institute, The University of Texas at Dallas, Richardson, Texas 75080, United States ‡ ERC, Inc., The Air Force Research Laboratory, Edwards AFB, California 93524-7680, United States § Department of Chemistry, University of Texas at Austin, Austin, Texas 78712, United States S Supporting Information *

ABSTRACT: A molecular switch that operates from the discrete expansion and contraction of helical polymer chain segments in response to chemical environment is studied and applied to volatile organic compound (VOC) sensing. Populations of expanded/contracted segments of poly(N-(1naphthyl)-N′-(n-octadecyl)carbodiimide (polyNOC) are found to be sensitive to solvent composition yet can be retained and modulated in the solid state. Previous studies have established that the two discrete local conformations of polyNOC can be measured in solution using infrared spectroscopy. Infrared spectroscopy, differential scanning calorimetry, and grazing incidence X-ray diffraction are used to study polyNOC in the solid state at various conformation populations. Results provide insight into the role of the core-dual shell structure of polyNOC on the mechanism of reversible switching between expanded/contracted, helical-rod conformations. Solid-state switching is applied to sensing VOC’s using infrared spectroscopy for transduction with the ability to distinguish between similar chemical compositions/structures, both polar and nonpolareven between n-hydrocarbons.

■

INTRODUCTION

Helical polycarbodiimides have been found to display chiroptical switching without isomerization of backbone atoms, without the presence of any obvious sensing group, and without stereogenic centers.21,22 The first polycarbodiimide to display reversible, chiroptical switching behavior was poly[N(1-anthracenyl)-N′-(n-octadecyl)carbodiimide], or polyAOC, displaying specific optical rotation values that could be modulated between SOR ≈ +300°/(dm/g)−1 at room temperature to 0°/(dm/g)−1 when dissolved in toluene and heated to ∼38 °C.23,24 A similar polycarbodiimide with 1naphthyl substituents instead of 1-anthracenyl, poly[N-(1naphthyl)-N′-(n-octadecyl)carbodiimide] (polyNOC; Figure 1), was later shown to display enhanced chiroptical switching performance (e.g., [SOR] ≈ +1360 at room temperature to [SOR]≈ −91° (g/mL)−1 cm−1 at 50 °C for toluene solutions) with the addition to hypersensitivity to solvent (e.g., polyNOC dissolved in DCM and chloroform displays [SOR] ≈ +1050 and −655° (g/mL)−1 cm−1, respectively).25,26 Structural changes associated with the chiroptical switching ability of polyNOC have been elucidated over the years through a combination of synthetic and computational efforts, including variable-temperature/solvent 15N NMR spectroscopic analysis of 15N-labeled polycarbodiimides and the calculation of

Biomimetic technologies that use the discrete conformational changes of individual synthetic macromolecules are emerging in areas of catalysis, microelectronics, biotechnology, and photovoltaics.1 Selective and nonselective sensors for VOCs are used for a variety of applications including food and beverage quality control to the diagnoses of disease and medical conditions.2−7 Despite the wide application of VOC sensing, commercially available technologies are mainly based on the collective properties of molecules for signaling (e.g., swelling, conductance, etc.).2−4,7−9 Individual chain technologies using discrete synthetic macromolecules possess the potential for achieving VOC sensing using simple transduction methods and data analysis. Helical polymers are commonly studied for their chiral properties10−13 but are increasingly used as sensing materials14−16 and molecular machines.17−19 Helical polymers are attractive as molecular sensors because they can undergo discrete changes in macromolecular conformation from weak interactions between pendant group moieties. The conformational switching process for helical polymers usually involves isomerization of backbone atoms in response to changes in steric and/or electronic character of pendant groups possessing sensing functional groups (e.g., H-bonding amides, π-stacking aromatics, etc.).12,14,16,20 © XXXX American Chemical Society

Received: May 26, 2017

A

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 1. Controlled, helix-sense selective polymerization of N-(1-naphthyl)-N′-(n-octadecyl)carbodiimide (NOC) using chiral (R)- or (S)-BINOL Ti(IV) diisopropoxide initiator forming the bistable, chiroptical switching polyNOC. Also shown are models (side view and down the helix) of the (P) and (M) helical polyNOC.

7-mer molecular models used to generate theoretical VCD spectra.27,28 NMR spectra for 15N-labeled polyNOC revealed that the chiroptical switching behavior is associated with equilibria between two discrete helical conformations. Variabletemperature 15N NMR data were used for van’t Hoff analysis, suggesting that the chiroptical switching process is mainly entropy driven and too low in energy (ΔG ≈ +0.3 kcal/mol) to be attributed to helix-sense inversion or isomerization of backbone atoms. VCD spectra of polyNOC in chloroform and DCM compared to simulated VCD spectra confirmed that the polycarbodiimide backbone does not undergo helical-sense inversion during chiroptical switching in dilute solutions and that the large optical rotation deviations are a result of a uniform flip of the pendant naphthyl rings from a right-handed to left-handed pendant helix. The VCD study also revealed that state A and state B possess different helical pitches, 5/1 and 7/ 2, respectively (Figure 2). The discrete conformations adopted by polyNOC helical segments, expanded (state A) and contracted (state B), produce separate absorbance signals in the imine stretching region of the infrared spectrum. Herein, solution and solid-state infrared spectroscopy are used to quantify the state A/state B

Figure 2. Structural characteristics of expanded/contracted, helical conformations adopted by (P)-polyNOC shown in two DFToptimized 7-mer models of states A and B. Black = polycarbodiimide backbone atoms; blue = pendant naphthyl substituents; and red = first carbon atom of octadecyl substituents.

population complimented by differential scanning calorimetry (DSC) and grazing-incidence X-ray diffraction (GI-XRD) to characterize n-octadecyl crystallization in the solid state. B

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 3. Kinetically trapping the conformational equilibrium of (M)-polyNOC solutions in the solid state as evidenced by transmission FTIR spectra of polyNOC solutions (left) and films cast onto KBr plates by spin-casting (middle) or drop-casting (right). Top to bottom: benzene, chloroform, and toluene (10 mg/mL).

■

RESULTS AND DISCUSSION Kinetically Trapping the State A/B Population of PolyNOC in the Solid State. Monitoring the equilibrium between state A and state B for polyNOC by IR spectroscopy allows for analysis in solution and the solid state. The population of state A/B remained constant in solution from 5 to 100 mg/mL (see Supporting Information section S1), suggesting that polyNOC intermolecular interactions do not influence the helical conformation of repeat units. This is attributed to the core−shell structure of polyNOC where densely grafted n-octadecyl substituents effectively shield the backbone from interacting with other polymer chains. The population of polyNOC adopted in solution could be kinetically trapped in both drop- and spin-casted films. Transmission FTIR spectra of polyNOC solutions and corresponding films prepared by either drop- or spin-casting onto KBr windows can be seen in Figure 3. The rate of evaporation influenced the final state A/state B population of polyNOC, which were monitored by measuring the overall peak height of the two vibrational modes at 1641 cm−1 (state B) and 1620 cm−1 (state A). Spin-cast films were closer to the population in solution than drop-cast film, most notably for state A enriched solutions (e.g., toluene; bottom of Figure 3). This trend was also seen with films drop- or sling-cast onto an ATR crystal (Figure S1). The population approaches a 0.8:1 ratio upon slow solvent removal, which is hypothesized to be the thermodynamic equilibrium in the absence of solvent, supported by variable temperature experiments discussed in a later section. Additional experiments were conducted to determine the role of n-octadecyl crystallization on the state A/B population. Crystallinity in PolyNOC Films with Different State A/ B Populations. Films drop-cast from different solvents produced free-standing films with physical property differences that could be seen with the naked eye (see Supporting Information section S2 for images and descriptions of polyNOC films drop-cast from different solvents). DSC and out-of-plane GI-WAXD experiments were conducted to

investigate how the state A/B population (i.e., casting solvent) influences crystallization behavior. The first temperature ramp of DSC experiments for (M)polyNOC films drop-cast from toluene, tetrahydrofuran, and benzene is shown in Figure 4. Films drop-cast from benzene (enriched with state B) displayed a relatively sharp endotherm centered at 120 °C (30.1 J/mol; 80.1 °C onset), consistent with a relatively ordered, single crystalline phase. Films drop-cast from toluene (enriched with state A) displayed broad, overlapping endotherms centered at ca. 105 and 110 °C and a smaller transition centered at ca. 50 °C (49.3 J/mol). Films drop-cast from tetrahydrofuran were almost completely amorphous, displaying two small endothermic transitions centered at 70 °C (0.2 J/mol) and 115 °C (0.3 J/mol). Thermograms of polyNOC films cast from benzene and toluene suggest a common melt associated with mainly intermolecular packing between n-octadecyl substituents with the addition of a broad collection of melting transitions for majority state A films. Keeping the densely grafted, core−dual shell structure of polyNOC in mind, additional crystalline phases associated with state A are consistent with intramolecular crystallization of confined n-octadecyl substituents in-between naphthyl substituents. The crystallization of the first few methylene units of n-octadecyl substituents sandwiched between rigid naphthyl substituents (i.e., inner shell) are consistent with an expanded helical pitch and would also explain why the state A/B population is stable in both solution and solid state−methylene units in the outer shell can be solvated or entangled without significantly influencing the backbone conformation or intramolecular crystallization within the inner shell. All polyNOC films were cycled from −90 to 150 °C after initial melting exhibiting an exotherm at ca. 20 °C that subsequently melted at ca. 10 °C when reheated (see Figure S4 for DSC data after the thermal history had been removed). The thermal transition behavior after the processing history had been removed is consistent with comb polymers possessing densely grafted n-octadecyl substituents and a rigid backbone. These comb polymers possess different melting points C

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 4. Overlay of DSC thermograms for (M)-polyNOC films (top left) drop-cast from toluene (dashed red trace), THF (solid black trace), and benzene (dotted blue trace). Traces correspond to the first heat from room temperature at 20 °C/min in a nitrogen environment. Bottom left: cartoon depiction and description of the polyNOC tertiary structure broken up into three parts: A rigid polycarbodiimide core and two shells that semicrystalline n-octadecyl substituents occupy. Top right: overlay of GI-WAXD patterns for (P)-polyNOC films drop-cast from DCM (red trace, middle), tetrahydrofuran (black trace, bottom), and chloroform (blue trace, top). Bottom right: side- (left) and end-view (right) of molecular models for two (P)-polyNOC chains (35 repeat units) in a fully state B conformation with n-octadecyl substituents fully extended in a zigzag conformation. The two polymer chains are shown at an overestimated maximum and minimum distance.

Table 1. Data Calculated from GI-XRD Patterns of (M)- and (P)-PolyNOC Drop-Cast from 25 mg/mL Solutions and the State A/B Population of Each Sample casting solvent (M)-polyNOC (M)-polyNOC (M)-polyNOC (P)-polyNOC (P)-polyNOC

DCM THF CHCl3 DCM CHCl3

d1, Å (rel int) 26.8 25.2 25.8 27.3 26.0

(100%) (100%) (100%) (100%) (100%)

d1, fwhm 0.69 0.83 0.83 0.68 0.83

depending on backbone rigidity: Tm for n-octadecyl substituents is highest when grafted to a semiflexible backbone (Tm ≈ 60 °C) and lowest for a rigid backbone (Tm ≈ 10 °C); flexible backbones have Tm ≈ 40 °C. The polycarbodiimide backbone is usually described as semiflexible, but the presence of bulky naphthyl substituents imbues rigidity, similar to a rigid poly(p-benzamide) backbone with n-octadecyl substituents consistent (Tm ≈ 10 °C), for example.29 Out-of-plane GI-WAXD measurements of polyNOC films drop-cast from different solvents displayed a correlation between the d-spacing and the state A/B population and a secondary diffraction peak that increases with state B content. Diffraction patterns of polyNOC films drop-cast from chloroform, tetrahydrofuran, or toluene are shown in Figure 4, upper right. Measurements and calculated distances from these patterns are summarized in Table 1 along with corresponding state A/B values. The principal diffraction peak ranged from 25 to 28 Å. This range is significantly lower than an estimated d1spacing from molecular models, ca. 30 Å, assuming that noctadecyl substituents are in a fully trans configuration and fully interdigitated (see Figure 4, lower right). The relatively short

d2, Å (rel int) 13.0 12.6 12.9 13.0 12.6

(15%) (26%) (27%) (16%) (32%)

d2, fwhm

state A/B (drop-cast)

0.76 0.84 0.81 0.74 0.77

1.4 1.0 0.8 1.4 0.8

distance for the crowded bottle-brush structure suggests a significant amount of entanglement between n-octadecyl substituents. The spacing between polyNOC chains, d1, increased with increasing state A content 2.5−2.6 nm for films drop-cast from benzene, tetrahydrofuran, and chloroform to between 2.7 and 2.8 nm when using DCM or toluene. The increase of d1 with increasing state A content is indicative of an overall decrease in n-octadecyl interdigitation between chains. Decreased interdigitation is consistent with intramolecular crystallization of noctadecyl substituents within the inner core of polyNOC in the expanded state A conformation. The secondary X-ray diffraction present in drop-cast polyNOC films, d2, decreased abundance with increasing state A content from 16% for films cast from DCM to 32% for films cast from chloroform. Additional XRD experiments are necessary to specifically define the crystalline phase changes associated with these polymer films. We hypothesize that the mechanism of thin-film switching may follow a similar solidstate transitions observed for helical poly(phenylacetylene)s D

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 5. Solvent-vapor response of (M)-polyNOC via dynamic ATR-FTIR. (M)-PolyNOC was either sling-cast from benzene (10 mg/mL) while being sequentially exposed to DCM (DCM) and chloroform (CHCl3) vapor (data on left side) or drop-cast from DCM (10 mg/mL ×3) and sequentially exposed to benzene and toluene vapor (data on right side). Top: state A/B vs time plot during the solvent exposure experiment with changes in chemical environment annotated with dashed, vertical lines and labels above plots (16-scan spectra were recorded every 10 s). Bottom, left: spectra of the imine stretching region initially (t = 0 min, black trace), after DCM exposure (t ≈ 4 min, dashed red trace), and after chloroform exposure (t ≈ 7 min, dotted blue trace). Bottom, right: spectra of the imine stretching region initially (t = 0 min, black trace), after toluene exposure (t ≈ 6 min, dashed red trace), and after benzene exposure (t ≈ 7 min, dotted blue trace).

(PPA’s) para-substituted with either n-octyl or 2-naphthyl substituents.30,31 The crystalline phases of n-octadecyl substituents in polyNOC films are assigned to broad, weak diffractions centered at about 15 and 18.5 2θ. The weak diffraction at 2θ ≈ 18.5 shifts to 2θ ≈ 20 for samples enriched with state A. Small-angle X-ray scattering (SAXS) measurements would be needed to identify crystalline phases associated with n-octadecyl substituents in state A and state B. Temperature Dependence of the State A/B Population. Once in the solid state, the population of states could be reversibly modulated with solvent vapor exposure or by heating/cooling beneath the melting point of n-octadecyl substituents. The influence of n-octadecyl crystallization on the state A/B population was probed by monitoring the state A/B population of polyNOC when heating/cooling above and below the melting point of n-octadecyl substituents. The state A/B population decreased as the temperature was raised from room temperature and could be partially reversed if cooled before reaching the melting point of n-octadecyl side chain, albeit attenuated compared to solution behavior (e.g., state A/B = 1.6 to 0.8 to 1.1 when heating/cooling from rt to 100 °C to rt, respectively). Once the melting point was reached the state A/B population would drop to state A/B = 0.8 and not change with subsequent cooling or heating. The fingerprint region of the infrared spectrum during heating/cooling suggested that phase transitions of n-octadecyl substituents occur with and without transitions between state A and B. (Data and detailed discussion for variable-temperature ATR-FTIR experiments are shown in Supporting Information section S5). Modulating the State A/B Population in PolyNOC Films with Solvent Vapor. The state A/B population of polyNOC was surprisingly responsive to solvent vapor

exposure. Simply placing a probe-type ATR-FTIR spectrometer crystal coated with polyNOC directly above a vial containing solvent produced a distinct change in the state A/B population, including switching between populations of majority state A to majority state B, and vice versa. A significant signal response would occur within the time between collecting 16-scan spectra (10 s between the start of each collection; ∼5 s for collecting 16 scans). The time for the state A/B population to equilibrate after solvent vapor exposure depended on film thickness and solvent volatility but would typically occur within 10−30 s. For example, the state A/B population of an (M)-polyNOC film while being successively exposed to DCM and chloroform vapor is presented in Figure 5, left side. Similar data for a thicker film of (M)-polyNOC being successively exposed to toluene and benzene vapor are shown on the right side of Figure 5. The “thin” (M)-polyNOC film was sling-cast from benzene displaying an initial state A/B population of 0.6. The state A/B population changed to 2.1 within 10 s of being exposed to DCM. The population fluctuated at extended exposure from the overlap of DCM IR signals. The population remained constant after the coated ATR-FTIR probe was removed from the solvent vapor. The state A/B population was then switched back to a majority state B population by exposure to chloroform vapor. The signal response occurred within 10 s and remained constant with extended exposure and after drying. The population was cycled back and forth between 2.1 and 1.9 and 0.6 with successive DCM and chloroform vapor exposures. The “thick” (M)-polyNOC film was drop-cast from DCM displaying an initial state A/B of 1.8. The conformation population would respond within 10 s of benzene vapor E

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

displayed in Figure 6, top). Nearly identical behavior was observed after exposing (M)-polyNOC to DCM vapor, drying in air, and then subsequently exposing to pentane and air again (second and third exposures displayed in Figure 6, bottom). Signal overlap from pentane CHC vibrational modes was initially suspected to account for the drop in state A content, however, exposing polyNOC in a majority state B condition to n-pentane produced an increase in state A during exposure that would then drop to slightly above the initial state A/B population after drying under nitrogen. The response to pentane vapor from a majority state B condition and was reproduced after subsequent exposures to n-heptane and noctane, suggesting that neither trapped solvent nor the initial population is associated with the anomalous response (Figure 6, top). The unique response of polyNOC to pentane vapor is suspected to be associated with molar volume and/or polarizability but requires further study. In contrast to n-pentane exposure, hexanes vapor produced a drop to state A/B = 0.5 from an initial majority state A condition that remained constant after vapor removal. This can be seen from the first exposure in the bottom plot of Figure 6 as well as in the top plot of Figure 6 after initially exposing to pentane. Similar behavior was also seen with n-heptane and noctane. If starting from a majority state A condition exposure to n-hydrocarbons produced a reproducible change to a specific population for each hydrocarbon analyte. Starting from a majority state B population, however, produced either no change or a variable change to a final state A/B value. The ability to distinguish between hydrocarbons implies that subtle physical interactions near the backbone of polyNOC can influence the delicate equilibria between expanded and contracted local conformations. This behavior is consistent with the intramolecular crystallization mechanism of switching described in this report. The regioregular primary structure of polyNOC, with alternating 1-naphthyl and n-octadecyl substituents, along a helical backbone with little/no helical reversals (i.e., defects in helix structure) produces a unique chemical environment in which confined methylene units can reproducibly and reversibly crystallize within the confined space between naphthyl substituents. The mechanism proposed herein explains why the helix-tohelix switching of polyNOC is fully reversible and hypersensitive to chemical environment. Solvents with Hansen solubility parameters closer to octadecyl substituents (i.e., polyethylene; σd = 17.10 J/cm3, σp = 3.10 J/cm3, σh = 5.2 J/ cm3) reduce the crystallinity of confined methylenes within the inner shell, increasing the population of the contracted, state B conformation. The state A/B population of polyNOC after contact with hydrocarbon vapor and their corresponding σd (σp = σh = 0 for hydrocarbons) are as follows: pentane σd = 14.5, state A/B = 0.66 ± 0.01; hexane σd = 14.9, state A/B = 0.49 ± 0.006, heptane σd = 15.3, state A/B = 0.44 ± 0.02; and octane σd = 15.5, state A/B = 0.45 ± 0.004.

exposure but would take 30−40 s to equilibrate (state A/B ca. 0.9), followed by no change after being dried in air. Subsequent exposure to toluene vapor produced an attenuated response reaching a population of state A/B ∼ 1.35 after 2.5 min of exposure that would drop to ∼1.3 after being dried. The process was repeated three times with little hysteresis in the state A/B population. Nonpolar VOC Sensing. The response of polyNOC to nonpolar hydrocarbons was reproduced several times under various conditions (i.e., different casting solvents, deposition methods, molecular weights, and helical sense). Dynamic ATRFTIR experiments designed to highlight the most interesting and nuanced response of (P)- and (M)-polyNOC films to hydrocarbons are shown in Figure 6. To begin, both samples

Figure 6. State A/B vs time plots for dynamic ATR-FTIR experiments (16-scan spectrum/10 s) of sling-cast films of (P)-polyNOC (top, 5 mg/mL toluene ×3) or (M)-PNOC (bottom, 5 mg/mL toluene ×3) while being exposed to hydrocarbon VOCs. The times where VOCs were introduced are annotated with colored, vertical lines with the analyte name shown above the plot; red delineates an increase in state A and blue a decrease.

exhibited the same state A/B populations when cast from the same solvents and exposed to the same VOCs, suggesting that the state A/B population is independent of helical sense. Consistent with the example shown in Figure 5, sling-cast polyNOC films displayed rapid state A/B population changes within 10 s of exposure to VOCs, equilibrating within 10−30 s. Unlike the previous examples, the conformation response was not always stable after the VOC analyte was removed and the film dried under a stream of nitrogen or in ambient conditions. This is highlighted in both of the experiments shown in Figure 6, most notably with exposure to n-pentane as well as with exposure to VOCs with similar initial state A/B populations (i.e., going from majority state B to majority state A). Exposure of n-pentane vapor to a DCM-cast film of (P)polyNOC produced a drop from state A/B = 1.9 to 1.0 would remain stable with extended exposure but would then equilibrate to state A/B = 2.0 when dried (first exposure

■

CONCLUSIONS The solid-state characterization of RR-PCDs that display chiroptical switching behavior in dilute solution has enabled a new method for the nonselective sensing of VOCs. The report described the facile assembly and use of a molecular sensor using commercial FTIR instruments (both transmission and ATR-FTIR). The change in state A/B population in response to VOCs was surprisingly rapid, responding, and equilibrating within 10−30 s. The imine stretching region has proven to be a F

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

support was provided by ERC, Inc., and the Air Force Office of Scientific Research (AFOSR).

convenient spectroscopic handle in which very few functional groups absorb allowing for sensing VOCs without signal occlusion or significant overlap. The promising results for the semiselective sensing of VOCs elicits further scrutiny for the development of a viable sensing method for industrial applications. Follow-up studies are underway focusing on correlating the state A/B population with solvent/polymer physical interactions (e.g., Hansen solubility parameters), developing an IR sensor array coated with different polycarbodiimide switches of tailored composition and structure, and potentially utilizing recent advances in synthesizing functional polycarbodiimides for controlling morphology32−34 to expand sensor function.

■

EXPERIMENTAL SECTION

■

ASSOCIATED CONTENT

■

Materials. (M)- and (P)-polyNOC were prepared using a literature procedure25 (scheme shown in the Introduction) with monomer:initiator values of 200:1 (Mn ∼ 84 kDa). FTIR Spectroscopy. Transmission FTIR spectra were collected on a Jasco Model 410 FTIR spectrometer either cast onto a KBr salt plate or in solution between KBr windows (0.11 mm path length). ATRFTIR spectra were collected on a Thermo Scientific Nicolet 380 ATRFTIR spectrometer. Dynamic ATR-FTIR spectra were collected using a probe-type, Mettler-Toledo reactIR ATR-FTIR spectrometer with a silicon sample window. Peak heights of imine stretches used to determine the state A/B population were normalized to stretches at either 776 or 1365 cm−1. Differential Scanning Calorimetry. Thermal transitions were characterized using a TA Q1500 differential scanning calorimeter. Samples were prepared by dissolving polyNOC at 10 mg/mL in toluene, THF, or benzene. The solutions were allowed to slowly evaporate in a saturated environment over the course of 17 h. The solutions were allowed to air-dry for several hours before being dried under vacuum overnight at room temperature. DSC experiments were conducted on 5 mg samples at 20 °C/min in a nitrogen environment. Grazing-Incidence, Wide-Angle X-ray Diffraction. Out-ofplane GI-XRD patterns were collected using a Shimadzu X-ray diffractometer equipped with a thin-film stage and Cu Kα source. Samples were drop-cast onto silicon wafers from 25 mg/mL solutions and allowed to slowly evaporate in an enclosed environment saturated with the same solvent, followed by drying at ambient. Samples were collected using the following instrument configuration and experimental parameters: cross-beam optics; parallel-beam geometry; incident angle (θ) = 0.5°; sampling step = 0.0100°; speed = 0.5 or 1°/min; attenuator = open; divergent slit (DS) = 0.2 mm; soller slit (SS) = open; and receiver slit (RS) = open.

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.7b01095. Figures S1−S6 (PDF)

■

REFERENCES

(1) Ouchi, M.; Badi, N.; Lutz, J.-F.; Sawamoto, M. Single-chain technology using discrete synthetic macromolecules. Nat. Chem. 2011, 3, 917−924. (2) Chiu, S.-W.; Tang, K.-T. Towards a chemiresistive sensorintegrated electronic nose: a review. Sensors 2013, 13, 14214−14247. (3) Smyth, H.; Cozzolino, D. Instrumental Methods (Spectroscopy, Electronic Nose, and Tongue) As Tools To Predict Taste and Aroma in Beverages: Advantages and Limitations. Chem. Rev. 2013, 113, 1429−1440. (4) Peris, M.; Escuder-Gilabert, L. On-line monitoring of food fermentation processes using electronic noses and electronic tongues: A review. Anal. Chim. Acta 2013, 804, 29−36. (5) Dragonieri, S.; van der Schee, M. P.; Massaro, T.; Schiavulli, N.; Brinkman, P.; Pinca, A.; et al. An electronic nose distinguishes exhaled breath of patients with Malignant Pleural Mesothelioma from controls. Lung Cancer 2012, 75, 326−331. (6) D’Amico, A.; Di Natale, C.; Falconi, C.; Martinelli, E.; Paolesse, R.; Pennazza, G.; et al. Detection and identification of cancers by the electronic nose. Expert Opin. Med. Diagn. 2012, 6, 175−185. (7) Roeck, F.; Barsan, N.; Weimar, U. Electronic Nose: Current Status and Future Trends. Chem. Rev. 2008, 108, 705−725. (8) Ryan, M. A.; Shevade, A. V.; Zhou, H.; Homer, M. L. Polymercarbon black composite sensors in an electronic nose for air-quality monitoring. MRS Bull. 2004, 29, 714−719. (9) Fonollosa, J.; Rodriguez-Lujan, I.; Shevade, A. V.; Homer, M. L.; Ryan, M. A.; Huerta, R. Human activity monitoring using gas sensor arrays. Sens. Actuators, B 2014, 199, 398−402. (10) Leiras, S.; Freire, F.; Quinoa, E.; Riguera, R. Reversible assembly of enantiomeric helical polymers: from fibers to gels. Chem. Sci. 2015, 6, 246−253. (11) Shimomura, K.; Ikai, T.; Kanoh, S.; Yashima, E.; Maeda, K. Switchable enantioseparation based on macromolecular memory of a helical polyacetylene in the solid state. Nat. Chem. 2014, 6 (5), 429− 434. (12) Yamamoto, T.; Adachi, T.; Suginome, M. Complementary Induction of Right- and Left-Handed Helical Structures by the Positioning of Chiral Groups on the Monomer Units: Introduction of (−)-Menthol as Side Chains of Poly(quinoxaline-2,3-diyl)s. ACS Macro Lett. 2013, 2, 790−793. (13) Ohta, E.; Sato, H.; Ando, S.; Kosaka, A.; Fukushima, T.; Hashizume, D.; et al. Redox-responsive molecular helices with highly condensed π-clouds. Nat. Chem. 2011, 3 (1), 68−73. (14) Nagata, Y.; Takagi, K.; Suginome, M. Solid Polymer Films Exhibiting Handedness-Switchable, Full-Color- Tunable Selective Reflection of Circularly Polarized Light. J. Am. Chem. Soc. 2014, 136, 9858−9861. (15) Bergueiro, J.; Freire, F.; Wendler, E. P.; Seco, J. M.; Quinoa, E.; Riguera, R. The ON/OFF switching by metal ions of the “Sergeants and Soldiers” chiral amplification effect on helical poly(phenylacetylene)s. Chem. Sci. 2014, 5 (6), 2170−2176. (16) Leiras, S.; Freire, F.; Seco, J. M.; Quinoa, E.; Riguera, R. Controlled modulation of the helical sense and the elongation of poly(phenylacetylene)s by polar and donor effects. Chem. Sci. 2013, 4 (7), 2735−2743. (17) Iamsaard, S.; Aßhoff, S. J.; Matt, B.; Kudernac, T.; Cornelissen, J. J. L. M.; Fletcher, S. P.; et al. Conversion of light into macroscopic helical motion. Nat. Chem. 2014, 6, 229−235. (18) Rudick, J. G.; Percec, V. Nanomechanical function made possible by suppressing structural transformations of polyarylacetylenes. Macromol. Chem. Phys. 2008, 209, 1759−1768. (19) Percec, V.; Rudick, J. G.; Peterca, M.; Heiney, P. A. Nanomechanical Function from Self-Organizable Dendronized Helical Polyphenylacetylenes. J. Am. Chem. Soc. 2008, 130, 7503−7508. (20) Okoshi, K.; Sakurai, S.; Ohsawa, S.; Kumaki, J.; Yashima, E. Control of main-chain stiffness of a helical poly(phenylacetylene) by

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Funding for this work was provided by the Faculty start-up fund from the University of Texas at Dallas (UTD) and the Endowed Chair for Excellence at UTD. Additional financial G

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules switching on and off the intramolecular hydrogen bonding through macromolecular helicity inversion. Angew. Chem., Int. Ed. 2006, 45, 8173−8176. (21) Kennemur, J. G.; Novak, B. M. Advances in polycarbodiimide chemistry. Polymer 2011, 52, 1693−1710. (22) Kennemur, J. G.; Novak, B. M. Hierarchical Chirality in Polycarbodiimides. Isr. J. Chem. 2011, 51, 1041−1051. (23) Tang, H.-Z.; Novak, B. M.; He, J.; Polavarapu, P. L. A thermal and solvocontrollable cylindrical nanoshutter based on a single screwsense helical polyguanidine. Angew. Chem., Int. Ed. 2005, 44, 7298− 7301. (24) Tang, H.-Z.; Boyle, P. D.; Novak, B. M. Chiroptical Switching Polyguanidine Synthesized by Helix-Sense-Selective Polymerization Using [(R)-3,3′-Dibromo-2,2′-binaphthoxy] (di-tert-butoxy)titanium (IV) Catalyst. J. Am. Chem. Soc. 2005, 127, 2136−2142. (25) Kennemur, J. G.; Clark, J. B.; Tian, G.; Novak, B. M. A New, More Versatile, Optical Switching Helical Polycarbodiimide Capable of Thermally Tuning Polarizations ± 359°. Macromolecules 2010, 43, 1867−1873. (26) Novak, B. M.; Tang, H.-Z., inventors; North Carolina State University. Devices containing polycarbodiimide polymers as chiroptical switching materials and methods of making polycarbodiimide polymers using optically active titanium alkoxide complex as polymerization catalyst. Patent US20070170422A1, 2007. (27) Merten, C.; Reuther, J. F.; DeSousa, J. D.; Novak, B. M. Identification of the specific, shutter-like conformational reorientation in a chiroptical switching polycarbodiimide by VCD spectroscopy. Phys. Chem. Chem. Phys. 2014, 16, 11456−11460. (28) Reuther, J. F.; Novak, B. M. Evidence of Entropy-Driven Bistability through 15N NMR Analysis of a Temperature- and SolventInduced, Chiroptical Switching Polycarbodiimide. J. Am. Chem. Soc. 2013, 135, 19292−19303. (29) Shi, H.; Zhao, Y.; Dong, X.; Zhou, Y.; Wang, D. Frustrated crystallisation and hierarchical self-assembly behaviour of comb-like polymers. Chem. Soc. Rev. 2013, 42, 2075−2099. (30) Motoshige, A.; Mawatari, Y.; Motoshige, Y. Y. R.; Tabata, M. Synthesis and solid state helix to helix rearrangement of poly(phenylacetylene) bearing n-octyl alkyl side chains. Polym. Chem. 2014, 5, 971−978. (31) Mawatari, Y.; Motoshige, A.; Yoshida, Y.; Motoshige, R.; Sasaki, T.; Tabata, M. Structural determination of stretched helix and contracted helix having yellow and red colors of poly(2-ethynylnaphthalene prepared with a [Rh(norbornadiene)Cl]2-triethylamine catalyst. Polymer 2014, 55, 2356−2361. (32) Reuther, J. F.; Siriwardane, D. A.; Campos, R.; Novak, B. M. Solvent Tunable Self-Assembly of Amphiphilic Rod-Coil Block Copolymers with Chiral, Helical Polycarbodiimide Segments: Polymeric Nano-structures with Variable Shapes and Sizes. Macromolecules 2015, 48 (19), 6890−6899. (33) Reuther, J. F.; Siriwardane, D. A.; Kulikov, O. V.; Batchelor, B. L.; Campos, R.; Novak, B. M. Facile Synthesis of Rod-Coil Block Copolymers with Chiral, Helical Polycarbodiimide Segments via Postpolymerization CuAAC “Click” Coupling of Functional End Groups. Macromolecules 2015, 48, 3207−3216. (34) Reuther, J. F.; Bhatt, M. P.; Tian, G.; Batchelor, B. L.; Campos, R.; Novak, B. M. Controlled Living Polymerization of Carbodiimides Using Versatile, Air-Stable Nickel(II) Initiators: Facile Incorporation of Helical, Rod-like Materials. Macromolecules 2014, 47, 4587−4595.

H

DOI: 10.1021/acs.macromol.7b01095 Macromolecules XXXX, XXX, XXX−XXX