Optical Rotatory Dispersion of Poly(l-lactic acid) (PLLA) in 19 Solvents and Study of PLLA Complexation with Polyphenylacetylene (PPA) in Solution

Abstract

Poly(l-lactic acid) or poly(l-lactide) (PLLA) is an optically active polymer derived from renewable sources and fully biodegradable. It is known that PLLA assumes a left-handed helix in the solid state and also in solution it still keeps a certain degree of helical structure. Here, we examine the Optical Rotatory Dispersion (ORD) behavior of two grades of PLLA (medium molecular weight and hexadecyl-terminated or high molecular weight for 3D printing) in 13 different solvents and analyze the experimental ORD data through the Moffitt–Yang equation. Furthermore, the ORD data of PLLA in additional 6 solvents were taken from the literature and analyzed with the Moffitt-Yang approach. The results suggest that, also in solution, PLLA maintains the left-handed helix, and the most structurizing and helicogenic solvents for PLLA are ethyl acetate, acetonitrile, and certain chlorinated solvents. The equilibrium association constant (K) and other thermodynamic parameters (ΔG°, ΔH° and ΔS°) between PLLA and polyphenylacetylene (PPA, another helical polymer in the solid state and in solution) were determined in trichloromethane, dichloromethane, and tetrahydrofuran. The K values found suggest a strong helix-helix interaction between the two polymers. The ORD analysis of the PLLA-PPA solutions shows evidence of the extrinsic Cotton effect and confirms the chiral helicity induction between the two polymers with 1:1 complex formation.

1. Introduction

Poly(l-lactic acid) (PLLA) or poly(lactide) is a unique biodegradable polymer produced from renewable raw materials on a large industrial scale [1,2,3]. Actually, PLLA finds many uses from packaging, automotive, and technical components, 3D printing, to medical devices, drug delivery, and cosmetics applications. PLLA is an intrinsically isotactic and stereoregular polymer with a helical structure in the solid state (see Figure 1) [1,2]. In more detail, PLLA exists in different crystal modifications. The most common consists of two different left-handed helix geometries having three units in one turn (3₁) in the β-form or ten units in three turns (10₃) in the most common α-form [1,2]. Stereocomplexes with remarkably higher melting temperatures than the homopolymers and considerably improved mechanical properties are obtained by intimate blending of PLLA and poly(d-lactic acid) (PDLA) [1,2,4]. The improved properties are explained by the presence of a dense network of weak CH⋯O=C hydrogen bonds [1,2,4].

Figure 1. Schematic structure of PLLA showing its isotacticity and the position of the asymmetric carbon atoms (with asterisk).

In this paper, our interest is addressed toward the PLLA behavior in solution in different solvents. Since 1965, there have been hints about the helical structure of PLLA in solution of certain solvents as suggested by the exploratory works from Schultz and Schwaab using optical rotatory dispersion (ORD) spectroscopy [5]. With some uncertainty [6], such hints about the PLLA helical structure in solution were confirmed a few years later [7]. Recently, such a finding was reached in a firm and conclusive way, through the breakthrough of vibrational circular dichroism (VCD) spectroscopy [8,9,10,11]. Indeed, PLLA assumes helical conformation in certain halogenated solvents as well as in certain polar solvents [8,9,10,11], confirming previous preliminary results obtained with optical rotatory dispersion (ORD) spectroscopy [5,6,7]. The macromolecular dynamics of PDLA and PLLA in dilute solutions were also studied by a number of different physical techniques, including small-angle X-ray scattering and static light scattering in a medium-polar solvent like tetrahydrofuran, revealing fine details of helical conformation [12,13]. Such exciting PLLA behavior in solution has led us to study with ORD its complex formation with iodine [14], C₆₀ and C₇₀ fullerenes [15,16], certain dyes [17], and during high-energy irradiation [18].

There is great interest in different types of PLLA complexes: with solvent molecules [19], in host-guest interactions, for example, with amylose [20,21], with different types of molecules like cyclodextrin, sorbitol, nanocellulose, humic substances, and biological molecules, either for composite production or for drug delivery purposes [22,23]. Furthermore, PLLA/gelatin and PLLA/collagen were developed for skin tissue engineering [24,25], and PLLA/chitosan composite for tissue regeneration [26]. Without any intent to be comprehensive, it is also worth mentioning the PLLA complexes and composites with poly(vinylacetate) [27,28], with poly(methyl methacrylate) [29,30,31], and with alginate [32].

The present work is first dedicated to a systematic study of the determination of PLLA helical content in 19 different solvents, analyzing the ORD results data through the Moffitt-Yang equation [33,34]. Furthermore, our attention will be focused on helix-helix interaction in solution between PLLA and polyphenylacetylene (PPA) with the determination of the equilibrium constant and other chemical thermodynamic parameters in selected solvents. In more detail, the PLLA-PPA interaction in solution will be studied by ORD with the detection of the extrinsic Cotton effect as proof of the chiral PLLA helices interaction and handedness induction on the PPA racemic helices [35,36,37,38]. In fact, it is known that phenylacetylene (and related monomers) undergoes a stereospecific polymerization over Rh(I) catalyst, yielding PPA (and other polyacetylenes) with helical structures in the solid state [39,40,41,42,43,44,45]. The PPA helical structure is also preserved in solution [43], and, depending on the solvent treatment, the heat or pressure treatment of PPA, it is possible to cause the cis-transoid to cis-cisoid helical structure transition (see Figure 2) [43,44,45]. The polyacetylenes (PAs) and polyphenylacetylenes (PPAs) synthesis, properties, and unique chiral properties have been reviewed numerous times [46,47,48]. Particularly interesting is the possibility to cause chiral induction in PPAs through non-covalent chiral interactions, including the interaction with solvents [49,50], and the switching of the handedness of the PPAs helices [51,52]. Regarding the PLLA-PPA interaction, little can be found in the literature. In one case, PPAs with chiral and racemic polylactide pendant groups were studied [53] while in another case, the helical chirality of racemic PAs was regulated by chiral polylactide in solid-state films to produce circularly polarized luminescence [54].

Figure 2. Stereoisomers of poly(phenylacetylene) (PPA).

2. Materials and Methods

2.1. Materials and Equipment

All solvents used in the present work were the best analytical grades purchased from Aldrich-Merck (Darmstadt, Germany). Two PLLA grades were used in the present work. A PLLA-3D, designed for 3D printing, was from Ingeo Biopolymers 2003D, produced by NatureWorks LLC (Blair, NE, USA) it was characterized by a molecular weight M_w = 182,000 Da and M_w/M_n ≈ 2 with DSC glass transition at +60 °C and by a DSC melting point peak at 157 °C with a melting enthalpy of 45.4 J/g [16,55]. A lower molecular weight PLLA with M_n ≈ 40,000 Da and hexadecyl-terminated was obtained from Aldrich-Merck (Darmstadt, Germany) and labelled PLLA-40kDa throughout this work. Further details about the properties of PLLA-40kDa can be found in the literature [56,57].

Phenylacetylene, the complex bicyclo[2.2.1]hepta-2,5-diene-rhodium(I) chloride dimer, i.e., [Rh(NBD)Cl]₂ and triethylamine (C₂H₅)₃N were purchased from Aldrich-Merck (Darmstadt, Germany).

Spectrophotometric studies were performed on a Shimadzu (Kyoto, Japan) UV-2450 spectrophotometer equipped with thermostated cells through the cell temperature controller TCC-240A. The optical rotatory dispersion (ORD) spectra were recorded on a Jasco (Japan) model P-2000 polarimeter equipped with a digital monochromator from Optometrics (Littleton, MA, USA), model DMC1-03, which transforms the polarimeter into a spectropolarimeter. A cell of 0.5 dm path length was used.

2.2. Preparation of the PLLA-3D or PLLA-40kDa Solutions

In Table 1 are summarized the PLLA-3D and PLLA-40kDa concentrations in the selected solvents used in the present study. In general, the PLLA concentration was kept at 0.5% (w/vol) with some exceptions. In a typical procedure, the PLLA was stirred at room temperature until its complete dissolution in the given solvent. Warming in a water bath was needed for the PLLA dissolution in ethyl acetate (AcOEt), acetone, toluene, chlorobenzene (Cl-Bz), and o-dichlorobenzene (o-DCBz). As also summarized in Table 1, some gel was noticed at the bottom of the flask of the cooled PLLA solution in the case of AcOEt, acetone, and toluene. In the case of o-DCBz, cooling causes the separation of the PLLA-3D from the solvent, and only warming above 50 °C ensures the complete dissolution. Thus, only for the o-DCBz solution was the ORD measurement made at 50 °C, while in all the other cases, the ORD measurements were made at ambient temperature.

2.3. Synthesis of Polyphenylacetylene (PPA)

Stereoregular PPA was synthesized using Rh(I) catalyst, the dimer complex with norbonadiene (NBD) complex, i.e., [Rh(NBD)Cl]₂ [39,40,41]. In detail, phenylacetylene (2.5 mL) was dissolved in 20 mL of toluene, and [Rh(NBD)Cl]₂ (0.7 mg) was added under stirring at room temperature. The reaction is initiated by the addition of a few drops of triethylamine to the stirred homogeneous solution. An immediate darkening of the solution was observed, as well as an increase in the viscosity. The mixture was left stirring overnight at ambient temperature, and then it was treated with a large excess of methanol. PPA precipitated as an orange mass, which was collected by filtration, washed repeatedly with methanol, and dried in air and in a desiccator to constant weight. The dried PPA was collected as an orange powder in nearly quantitative yield.

2.4. Determination of the Equilibrium Constant Between PLLA-40kDa and PPA

The equilibrium constant in the PLLA-PPA interaction in solution was determined with the same general procedure used in the previous works [14,15,16,58,59]. A stock solution of PPA (1.70 mg) in 100 mL of tetrahydrofuran (THF), dichloromethane (CH₂Cl₂), or chloroform (CHCl₃) was prepared. The electronic absorption spectrum of the solution was recorded at a selected temperature using the thermostated cell holder of the spectrophotometer. Then PLLA-40kDa was added step-by-step in weighted amounts to the stock solution. After each PLLA-40kDa addition and dissolution, the resulting electronic spectrum was recorded. PLLA-40kDa was selected for its prompt solubility in the selected solvents. The spectrophotometric experimental data were then treated according to Scott’s equation [60] for the determination of K_eq at a given temperature. Through the Van’t Hoff approach, it was then possible to derive other thermodynamic parameters regarding the PLLA-PPA interaction in solution.

3. Results and Discussion

3.1. Analysis of the ORD Data with the Moffitt-Yang Equation

3.1.1. ORD Measurement of PLLA in Selected Solvents

In this study, the ordinary PLLA commonly used was a version designed for 3D printing and hence referred to hereinafter as PLLA-3D or simply 3D. It is characterized by high molecular weight and high strength (see Section 2) [55]. In some specific cases, a low molecular weight PLLA, referred to as PLLA-40kDa or simply 40kDa, was used. The latter polymer was also characterized by chain termination with a long hexadecyl aliphatic chain, ensuring a quick and complete solubility in certain solvents like chloroform, dichloromethane, and tetrahydrofuran (for further details regarding PLLA-40kDa see Section 2) [56,57].

The ORD measurement consists of the measurement of the degree of rotation of the plane of polarized light α crossing a solution of concentration c (in g/mL) kept in a cell with l (in dm) path length by sweeping the wavelength λ, so that the specific optical rotation [α]_λ is determined [34,35]:

[α]_λ = α l⁻¹ c⁻¹

(1)

Figure 3 shows the ORD of the two PLLA grades, selected for this work, dissolved in chloroform, dichloromethane, and N-methyl-2-pyrrolidone (NMP). In all these three solvents, PLLA-40kDa shows a significantly more pronounced specific optical rotation than the higher molecular weight PLLA-3D. Thus, the lower molecular weight PLLA-40kDa with the hexadecyl end group appears to offer more structured solutions than the high molecular weight type. These results will be discussed more deeply after the analysis of the ORD data according to the Moffitt–Yang equation, in the next section.

Figure 3. ORD of PLLA-3D (circles) and PLLA-40kDa (triangles) at ambient temperature in three different solvents: N-methyl-2-pyrrolidone (NMP, green traces); dichloromethane (blue traces), and chloroform (red traces).

In addition to the ORD comparison of PLLA-40kDa versus PLLA-3D in three selected solvents (Figure 3), the latter polymer was studied in 13 different solvents as shown in Figure 4. In Figure 4a are reported the ORD curves of PLLA-3D in a series of chlorinated solvents (chloroform, 1,1’,2,2’-tetrachloroethane, dichloromethane and chlorobenzene) and two ethers (tetrahydrofuran and 1,4-dioxane). All the measurements were made at ambient temperature. Additionally, PLLA-3D in o-dichlorobenzene was studied with ORD (not shown in Figure 4a for clarity), but the ORD curve, only in this case, was collected at 50 °C because at lower temperatures, the PLLA-3D precipitates from the solution. Anyway, the ORD of PLLA-3D in o-dichlorobenzene resembles that in chlorobenzene, as will be discussed in the next section.

Figure 4. ORD of PLLA-3D at the ambient temperature (a) in chlorinated solvents and in ethers: CHCl₃ (blue), Cl₂CH-CHCl₂ (TCE, orange); tetrahydrofuran (THF, grey), 1,4-dioxane (red), CH₂Cl₂ (green), chlorobenzene (black) (b) in various other solvents: CH₃CN (violet), ethyl acetate (AcOEt, red), acetone (orange), N,N’-dimethylformamide (DMF, black), toluene (blue), N-methyl-2-pyrrolidone (NMP, green).

Figure 4b shows the ORD curves of PLLA-3D in polar aprotic solvents like acetonitrile, ethyl acetate, acetone, N,N’-dimethylformamide and N-methyl-2-pyrrolidone. Furthermore, the PLLA-3D ORD curve in apolar toluene is also reported.

At first glance, it is immediately evident from Figure 4a that at approximate same concentration and temperature conditions, PLLA-3D is giving significantly more negative optical rotation in chlorobenzene than in 1,1’,2,2’-tetrachlorobenzene. Similarly, in Figure 4b, it is immediately evident that PLLA-3D is showing much higher negative rotatory power in ethyl acetate and in acetonitrile than in acetone or in N-methyl-2-pyrrolidone. Thus, higher rotatory power implies that the PLLA-3D is assuming a helical structure in certain solvents in a way more pronounced than in other solvents. In other words, there are solvents that favor the formation and retention of the helical structure of PLLA-3D it has in the solid state, i.e., left-handed helix geometries having three units in one turn (3₁) or ten units in three turns (10₃) [1,2], while other solvents are denaturing the helical structure, promoting at least in part the emergence of a statistical coil. The solvent favoring the helical structure are also known as helicogenic solvents.

To analyze in a deeper way the solvent effect on the PLLA conformation in different solvents, the ORD data were analyzed according to the Moffitt model through the Moffitt-Yang equation.

3.1.2. The Moffitt Model and the Moffitt-Yang Equation

In 1956, Moffitt proposed a model to describe the optical activity of helical polymers in solution [61]. The model is based on the interaction of transition dipoles within a helical array. More in detail, Moffitt proposed that the electronic transition occurring in each monomeric unit (the ketone or ester group of each lactic acid residue in PLLA) should be split into two components, one parallel and one perpendicular to the helix axis. This approach is known as exciton splitting, and the model assumes an infinitely long and uniform helix. With our modern view of polymer chains in solution, it is immediately evident that the Moffitt model involves a considerable oversimplification of the actual physical situation [62,63]. Nevertheless, the Moffitt-Yang equation [34,35,63] remains a powerful phenomenological equation for the analysis of the ORD data and for the estimation of the helical content of the optically active polymers.

The Moffitt–Yang equation reports the specific optical rotation [α] in terms of mean residue rotation [m’]_λ [34,35]:

[m’]_λ = {3(n² − 2)⁻¹}{(M_m [α])/100}

(2)

The first term of Equation (1), i.e., {3(n² − 2)⁻¹} is the Lorentz refractive index correction for the selected solvent (see Table A1 reported in the Appendix A for the refractive index values of each solvent and other solvent properties) with n the refractive index of solvent in the spectral window considered for the ORD measurement (in practice it is used the n_D of the solvent), M_m is the molecular weight of the monomeric unit of the polymer. In the case of PLLA, the lactic acid residue has a M_m = 72.06 Da. The specific optical rotation [α]_λ was already defined by Equation (1).

After these premises, the Moffitt–Yang equation is reported as [34,35]:

[m’]_λ = a₀ [λ₀²/(λ² − λ₀²)] + b₀ [λ₀⁴/(λ² − λ₀²)²]

(3)

with

a₀ a parameter related to the intrinsic optical rotation of the monomeric units and random coil together with the solvent effect;

b₀ is the parameter directly related to the helical content in the given solvent;

λ₀ is a constant that linearizes the plot and is taken in correspondence with the electronic transition of the optically active chromophore [34,35]. In the case of polypeptides, λ₀ is taken at 212 nm in correspondence to the amide electronic transition [34,35], while was taken at 201 nm for PLLA [5]. Our recent study on PLLA thin solid film has shown that the electronic transition due to the ester group occurs at 209.5 nm [16]. Thus, instead of adopting a λ₀ from literature, for our analysis of the ORD data on PLLA in various solvents, λ₀ = 209.5 nm has been our choice. After all, the n → π* transition of the C=O group in simple esters dissolved in hydrocarbons is found just at 210–211 nm [64].

To determine a₀ and b₀ parameters, the preferred plottable version of Equation (3) is under the form of [34,35]:

[m’]_λ [(λ² − λ₀²)/λ₀²] = a₀ + b₀ [λ₀²/(λ² − λ₀²)]

(4)

So that b₀ is determined from the slope and a₀ from the intercept. The physical dimensions of both a₀ and b₀ are the same as the residue rotation [m’] expressed in (deg cm²)/(dmol).

A couple of examples of the analysis of the ORD data with Equation (4) are provided in Figure 5. The ORD data of PLLA-3D in CH₃CN and in CHCl₃ were plotted according to Equation 4. From the slope b₀ = 349.0 in CH₃CN and 307.1 in CHCl₃. From the intercept a₀ = −698.1 and −542.3, respectively. A similar treatment (according to Equation (4)) was applied to all the ORD data in 13 different solvents and 2 PLLA grades. The results are collected in Table 1, where all solvents are listed according to decreasing values of b₀. The solvents with high b₀ values favor the helical conformation of PLLA in solution, and this is the case, for instance, of ethyl acetate, acetonitrile, o-dichlorobenzene, and chlorobenzene. The worst solvents in terms of helical content appear to be 1,1’,2,2’-tetrachloroethane and acetone, characterized by the lowest b₀ values.

Figure 5. Moffitt–Yang plot according to Equation (4) of the ORD data of PLLA-3D in CH₃CN (orange dots) and in CHCl₃ (blue dots). In CH₃CN b₀ = 349.0 from the slope and a₀ = −698.1 from the intercept while in CHCl3 b₀ = 307.1 and a₀ = −542.3 respectively.

Another important aspect reported in Table 1 regards the positive values of b₀ in all solvents studied. This is in line with earlier results and confirms that PLLA assumes a left-handed helix also in solution [5,6,7]. In Figure 6, the expected linear correlation of the Specific Optical Rotation [α]_λ with b₀ and a₀ Moffitt-Yang parameters using the data of Table 1 is shown. The specific optical rotation of PLLA in any solvent, measured for example at 380 nm, is linearly correlated with the b₀ parameter: [α]₃₈₀ = −1.0582 b₀, as shown in Figure 6a. Indeed, a higher degree of helical conformation in solution gives an extra optical rotatory power. The linear correlation is also very good with the a₀ parameter, e.g., [α]₃₈₀ = 0.5850 a₀; in this case, a₀ is linked first of all to the intrinsic optical activity of the monomeric units, but it is also expressing the solvent interaction with the polymer chains [34,35]. In other words, there are two contributions to a₀, the monomer effect, i.e., the intrinsic chirality of the l-lactide units, and the solvent effect, which is a result of the PLLA interaction with solvent dipoles and polarizability. All the a₀ values in Table 1 are negative values, it confirms that the l-configuration of the l-lactide units is dominating the non-helical part of the ORD signal. Of course, there is also a linear correlation between the b₀ and a₀ values of Table 1.

Figure 6. Correlation of the specific optical rotation [α]₅₈₉ and [α]₃₈₀ with the Moffitt–Yang b₀ and a₀ parameters; (a) [α]₅₈₉ = −0.4644 b₀ (blue dots) and [α]₃₈₀ = −1.0582 b₀ (orange dots) (b) [α]₅₈₉ = 0.2565 a₀ (blue dots) and [α]₃₈₀ = 0.5850 a₀ (orange dots).

The anomaly in Table 1 is represented by the solvent N-methyl-pyrrolidin-2-one (NMP), which shows a considerably high b₀ value, suggesting a high helical content, but this is not corroborated by the specific optical rotation values, which are unusually low, comparable to those found in solvents like TCE and acetone, which disfavour the helix conformation of PLLA in solution.

For this reason, the Moffitt-Yang parameters in NMP solvent are shown separately in Table 1 and were not included in the correlation graphs of Figure 6. It is necessary to clarify that the ORD data in NMP solutions were collected in a restricted spectral range of 600–390 nm instead of the usual spectral range of 620–350 nm adopted for all other PLLA solutions. However, in our view, the anomalous behavior of NMP should be attributed mainly to a strong polymer-solvent interaction, since NMP is the solvent with the highest dipole moment (4.09 Debye) among all solvents studied in Table 1 [65]. The anomaly of the NMP solvent can be explained in terms of PLLA-NMP complex formation, a phenomenon already observed with highly polar aprotic solvents [19].

Table 1. Moffitt–Yang b₀ and a₀ parameters, specific optical rotations, and concentration of PLLA in selected solvents at ambient temperature. Polarity and polarizability of solvents.

SOLVENT and SOLUTE	SOLVENT POLARITY (ε)/(μ) (*)	Polarizability (**)	b0 (Equation (4))	a0 (Equation (4))	[α]589	[α]380	% (w/V) (***)
Ethyl Acetate (****) [PLLA-3D]	6.02/2.05	0.3061	411.1	−722.7	−176.1	−401.8	0.135
Acetonitrile [PLLA-3D]	37.5/3.39	0.2875	349.0	−698.1	−168.7	−396.1	0.484
o-Dichlorobenzene (†) [PLLA-3D]	9.93/2.56	0.4118	345.8	−571.7	−159.3	−341.4	0.246
Chlorobenzene [PLLA-3D]	5.71/1.75	0.3981	337.9	−599.8	−165.3	−367.2	0.246
Chloroform [PLLA-40kDa]	4.81/1.10	0.3529	328.9	−585.8	−149.5	−344.9	0.670
Dichloromethane [PLLA-40kDa]	9.08/1.85	0.3396	325.5	−593.6	−150.7	−345.8	0.450
Toluene (****) [PLLA-3D]	2.38/0.36	0.3929	324.7	−587.0	−157.9	−359.3	0.450
Dioxane [PLLA-3D]	2.21/0.0	0.3382	314.6	−567.1	−144.7	−329.2	0.484
Chloroform [PLLA-3D]	4.81/1.1	0.3529	307.1	−542.3	−138.7	−318.6	0.670
N,N’-dimethylformamide [PLLA-3D]	36.7/3.85	0.3435	306.9	−560.7	−142.6	−328.8	0.463
Tetrahydrofuran [PLLA-3D)	8.20/1.63	0.3290	303.9	−552.8	−138.9	−319.7	0.495
Dichloromethane [PLLA-3D]	9.08/1.85	0.3396	300.7	−554.6	−141.4	−323.5	0.423
1,1’,2,2’-Tetrachloroethane [PLLA-3D]	8.2/1.85	0.3812	247.4	−487.5	−134.6	−303.5	0.413
Acetone (****) [PLLA-3D]	20.7/2.86	0.2973	240.4	−453.4	−106.9	−252.6	0.467
N-Methyl-pyrrolidin-2-one (‡) [PLLA-40kDa]	32.3/4.09	0.3673	407.0	−508.8	−127.9	−274.7	0.261
N-Methyl-pyrrolidin-2-one (‡) [PLLA-3D]	32.3/4.09	0.3673	363.2	−481.3	−122.2	−266.2	0.468

(*) ε = dielectric permittivity [66]; μ = dipole moment in Debye [65]; (**) Onsager solvent polarizability (n² − 1)/(2n² + 1); (***) concentration g/100 mL; (****) some gel present; (†) ORD taken at T ≈ 50 °C; (‡) ORD in a restricted spectral range of 600–380 nm instead of the usual 620–350 nm for all the other solvents. General note: for further solvent properties and descriptors, see Table A1 in the Appendix.

Table 1. Moffitt–Yang b₀ and a₀ parameters, specific optical rotations, and concentration of PLLA in selected solvents at ambient temperature. Polarity and polarizability of solvents.

SOLVENT and SOLUTE	SOLVENT POLARITY (ε)/(μ) (*)	Polarizability (**)	b0 (Equation (4))	a0 (Equation (4))	[α]589	[α]380	% (w/V) (***)
Ethyl Acetate (****) [PLLA-3D]	6.02/2.05	0.3061	411.1	−722.7	−176.1	−401.8	0.135
Acetonitrile [PLLA-3D]	37.5/3.39	0.2875	349.0	−698.1	−168.7	−396.1	0.484
o-Dichlorobenzene (†) [PLLA-3D]	9.93/2.56	0.4118	345.8	−571.7	−159.3	−341.4	0.246
Chlorobenzene [PLLA-3D]	5.71/1.75	0.3981	337.9	−599.8	−165.3	−367.2	0.246
Chloroform [PLLA-40kDa]	4.81/1.10	0.3529	328.9	−585.8	−149.5	−344.9	0.670
Dichloromethane [PLLA-40kDa]	9.08/1.85	0.3396	325.5	−593.6	−150.7	−345.8	0.450
Toluene (****) [PLLA-3D]	2.38/0.36	0.3929	324.7	−587.0	−157.9	−359.3	0.450
Dioxane [PLLA-3D]	2.21/0.0	0.3382	314.6	−567.1	−144.7	−329.2	0.484
Chloroform [PLLA-3D]	4.81/1.1	0.3529	307.1	−542.3	−138.7	−318.6	0.670
N,N’-dimethylformamide [PLLA-3D]	36.7/3.85	0.3435	306.9	−560.7	−142.6	−328.8	0.463
Tetrahydrofuran [PLLA-3D)	8.20/1.63	0.3290	303.9	−552.8	−138.9	−319.7	0.495
Dichloromethane [PLLA-3D]	9.08/1.85	0.3396	300.7	−554.6	−141.4	−323.5	0.423
1,1’,2,2’-Tetrachloroethane [PLLA-3D]	8.2/1.85	0.3812	247.4	−487.5	−134.6	−303.5	0.413
Acetone (****) [PLLA-3D]	20.7/2.86	0.2973	240.4	−453.4	−106.9	−252.6	0.467
N-Methyl-pyrrolidin-2-one (‡) [PLLA-40kDa]	32.3/4.09	0.3673	407.0	−508.8	−127.9	−274.7	0.261
N-Methyl-pyrrolidin-2-one (‡) [PLLA-3D]	32.3/4.09	0.3673	363.2	−481.3	−122.2	−266.2	0.468

(*) ε = dielectric permittivity [66]; μ = dipole moment in Debye [65]; (**) Onsager solvent polarizability (n² − 1)/(2n² + 1); (***) concentration g/100 mL; (****) some gel present; (†) ORD taken at T ≈ 50 °C; (‡) ORD in a restricted spectral range of 600–380 nm instead of the usual 620–350 nm for all the other solvents. General note: for further solvent properties and descriptors, see Table A1 in the Appendix.

In this context, to complete the discussion, it is necessary to recall that for the PLLA helix to exist, the torsion angles along the polymer backbone must fall into a specific pattern: the gt (gauche-trans) conformation, which refers to the rotational state of the -(O=C)-O-C_α-(C=O)- and -O-C_α-(C=O)-O- sequences in a crystalline ordered phase. In solution, of course, the bonds are free to rotate, producing various conformers (e.g., gg, gt, tg, tt). “Helical fraction” of PLLA in solution should not be viewed as long, rigid rods (this is the limitation of the Moffitt model).

Instead, it should be conceived as the statistical probability that segments of the chain are currently assuming the gt arrangement. Thus, it is correct to talk about a population of gt conformers to express the helical fraction of a polymer in solution. The solvent chemical nature and polarity are able to greatly affect such a population of gt conformers. In PLLA solutions, solvent polarity acts as a “fine-tuning knob” for the population of gt conformers. Because the PLLA 10₃ helix relies on specific dipole alignments and steric packing, the solvent can either stabilize these building blocks or disrupt them into a random coil. The PLLA helix has a net dipole moment directed along its axis. In a less polar solvent, the polymer chain “collapses” into an ordered helical structure to internally satisfy its own dipole-dipole interactions. In polar solvents (like DMF or NMP), the PLLA ester groups can interact directly. This competitive interaction often “pulls” the bonds away from the gt state into a disordered random coil (e.g., gg and tt conformers) because the solvent molecules are more prone to solvating individual monomeric units than letting the chain wrap around itself. Interestingly, some specific polar solvents don’t just disrupt the helix—they participate in it. Solvents like THF or DMF can form “complex crystals” (ε-form) with PLLA at low temperatures [19]. In these cases, the solvent molecules nestle into the helical voids, actually stabilizing the gt conformers into a unique 10₃ helix that wouldn’t be as stable in a purely inert environment.

While no simple correlations were found between the b₀ and a₀ values with the dielectric permittivity of the solvents or the dipole moment (reported μ values in Table 1) it was instead found that solvent polarizability (or Onsager function of the refractive index):

(n² − 1)/(2n² + 1)

(5)

shows a quite good correlation with both b₀ and a₀ Moffitt–Yang parameters, as shown in Figure 7. This correlation suggests that it is not only a matter of dipole-dipole interaction since solvent polarizability also acts as a stabilizer for the helical architecture in solution. This linear correlation with the Onsager function [F(n²)] is strong evidence that the helix formation is also driven by dispersion-driven solvation rather than just simple dielectric effects. As the solvent becomes more polarizable (higher F(n²)), the dispersion forces and induced-dipole interactions between the solvent and the PLLA backbone strengthen. These interactions favor the gt conformer. The solvent essentially stabilizes the specific geometry of the 10₃ helix, leading to a higher population of helical segments. In PLLA’s specific optical configuration, this manifests as an increase in the positive b₀ value.

Figure 7. Correlation of the Moffitt–Yang parameters b₀ and a₀ with the Onsager function of the solvent polarizability.

The a₀ parameter reflects the “intrinsic” optical activity of the monomer units and the non-helical contributions. The fact that a₀ becomes more negative as polarizability F(n²) increases suggests a strong electronic effect of the solvent on the ester chromophores. As the solvent’s electron cloud becomes more “distortable” (higher F(n²) values), it interacts more intensely with the n → π* electronic transitions of the lactic acid units. This shift in the local electronic environment of the monomer pushes the a₀ value further into the negative, likely due to a change in the local solvation shell as the chain transitions from a coil to a more ordered helical state.

Correlations were also attempted between b₀ and a₀ values and the Reichardt (E_T(30) or Nile Red (NR) polarity scales and other solvent properties and descriptors reported in Table A1 of the Appendix. However, no satisfactory correlation was achieved.

In Figure 3 and in Table 1, it is also evidenced that the behavior of the two PLLA grades studied in the same solvents and at the same concentration and temperature. The b₀ value of PLLA-40kDa is systematically higher than that of the conventional PLLA-3D. In CHCl₃, b₀ is 328.9 vs. 307.1, respectively, in CH₂Cl_2, 325.5 vs. 300.7, and even in NMP b₀ = 407.0 vs. 363.2. Because of its lower molecular weight and long hexadecyl chain end group, PLLA-40kDa shows a ready solubility in various solvents like the mentioned CHCl₃ and CH₂Cl_2, but also THF, and a higher trend to produce a population of gt conformers leading to a higher helical fraction with respect to PLLA-3D in solution. The lower molecular weight PLLA-40kDa is characterized by higher chain mobility in solution, allowing the chains to rearrange into helical “pre-ordered” structures more rapidly. On the other hand, in the case of the higher PLLA-3D, the high degree of entanglement acts as a physical barrier that can “trap” chains in a disordered state, making it harder for long-range helical order to develop quickly.

The long alkyl chain end group is believed to act as a lubricant in the PLLA-40kDa, acting like a plasticizer and reducing the glass transition temperature of the polymer by increasing the free volume between the polymer chains [56,57]. PLLA-40kDa will show a high concentration of individual, solvated helices. The hexadecyl tail acts as a spacer, further reducing the chance of these helices aggregating compared to an acid-terminated chain. The differences between PLLA-3D and PLLA-40kDa can be summarized in the fact that the former relies on chain entanglement for its properties, whereas the 40kDa ester-terminated version relies on end-group chemistry [56,57].

3.1.3. The Moffitt–Yang Equation Applied to Previous ORD Data: A Comparison with the Current Data

In Table 1 are reported the ORD data of PLLA (two grades) solutions in 13 different solvents analyzed with the Moffitt-Yang equation. In the scientific literature, the ORD data of PLLA dissolved in 9 solvents [5,6] are also available, 6 of which are not reported in Table 1. These data were analyzed according to Equation 4, and the results are shown in Table 2. Before commenting on the results in Table 2, it is necessary to clarify that the PLLA from the literature is different from the two grades studied in the present work, in terms of molecular weight and end groups. Furthermore, the spectral range of the ORD data is also different. In our current work, the spectral range of the ORD data is comprised between 350 and 620 nm, with λ₀ selected at 209.5 nm, which is the absorption maximum of the PLLA-3D thin solid film [16]. The spectral range of the ORD data of the PLLA solutions in ref. [5] is instead restricted in the 367–578 nm range. In order not to alter the original b₀ and a₀ values found by these authors [5], the λ₀ = 201 nm chosen by them for the Moffitt-Yang analysis of these older ORD data was maintained. In the case of ref. [6], the ORD spectra range considered is shifted to shorter wavelengths, i.e., 200–400 nm. However, the Moffitt-Yang analysis of the ORD data requires considering only the values far away from the electronic transition of the chromophore [34,35]. Consequently, in our Moffitt–Yang analysis of the old ORD data of ref. [6], we have used only the 280–400 nm spectral range, and for λ₀ we have adopted 209.5 nm, i.e., the same value used for the analysis of the solvents listed in Table 1 and due to the PLLA electronic transition in the solid state as a thin solid film [16].

Table 2. Moffitt–Yang b₀ and a₀ parameters, recalculated from refs. [5,6] and compared with the current work.

SOLVENT (‡)	b0 (Equation (4))	a0 (Equation (4))	λ0 (nm) (*)	Spectral Range (nm) (**)	% (w/V) (***)	Reference
Acetonitrile PLLA-3D	349.0	−698.1	209.5	350–620	0.484	This Work
Acetonitrile	277.3	−572.9	209.5	280–400	n.r. (#)	[6]
Chloroform PLLA-40kDa	328.9	−585.8	209.5	350–620	0.67	This Work
Chloroform	317.3	−634.4	201.0	365–578	0.72	[5]
Chloroform PLLA-3D	307.1	−542.3	209.5	350–620	0.67	This Work
Chloroform	280.9	−713.2	209.5	280–400	n.r. (#)	[6]
Dichloromethane PLLA-40kDa	325.5	−593.6	209.5	350–620	0.45	This Work
Dichloromethane	309.9	−654.8	201.0	365–578	0.72	[5]
Dichloromethane PLLA-3D	300.7	−554.6	209.5	350–620	0.42	This Work
m-Cresol	305.7	−984.6	201.0	365–578	0.72	[5]
Trifluoroethanol	298.8	−899.9	209.5	280–400	n.r. (#)	[6]
1,1,2-Trichloroethane	297.3	−574.3	201.0	365–578	0.72	[5]
Dichloroacetic acid	279.7	−680.9	201.0	365–578	0.72	[5]
Trifluoroacetic acid	273.9	−901.8	209.5	260–400	n.r. (#)	[6]
Sulfuric acid	199.6	−585.3	201.0	365–578	0.72	[5]

(*) λ₀ used in the Equation (4); (**) spectral range covered in the ORD measurements; (***) concentration g/100 mL; (‡) The PLLA grade is reported only in the measurements of the present work; (^#) n.r. = not reported. General note: for further solvents properties and descriptors, see Table A1 in the Appendix.

Because of the just-mentioned limitations, it is immediately evident from Table 2 that the PLLA-3D in CH₃CN yields b₀ = 349.0 in the present work, a value significantly higher than b₀ = 277.3 derived from older data [6]. Relatively little is known about the PLLA used by the authors of ref. [6] and even its concentration in solution was not specified. Furthermore, the difference in the spectral range (see Table 2) may have led to such important differences. Regarding the CHCl₃ solvent, in Table 2 we have available 4 different PLLA grades and, as expected, the largest b₀ value is offered again by PLLA-40kDa, with a value not far from that shown by the PLLA of ref. [5]. Also, the PLLA-3D offers a b₀ value reasonably close to the previous two values. Completely offset results the b₀ value the PLLA used in ref. [6], in the same solvent. Here, it is necessary to apply the same explanations given for the b₀ value differences in CH₃CN. Exactly the same b₀ sequence found in chloroform can be found in CH₂Cl₂ solvent in Table 2: PLLA-40kDa > PLLA [5] > PLLA-3D. This is not a surprise, considering the similar chemical nature of the two chlorinated solvents, and confirms the consistency of the results.

Table 2 shows also the Moffitt–Yang parameters of an additional 6 solvents not considered in the present work (Table 1) but taken from literature ORD data [5,6]. Five of these solvents are acidic and highly polar solvents and, in general, give b₀ < 300, which suggests a trend of moderate to lower and lower PLLA helical structure induction in these solvents on passing from m-cresol to sulfuric acid, following the sequence reported in Table 2. Regarding the fluorinated solvents in Table 2, it is worth reminding that trifluoroethanol is considered a helix-inducing solvent in protein chemistry [35], and also in Table 2, with PLLA, its b₀ ≈ 300 suggests a similar behavior. Instead, trifluoroacetic acid is definitely considered a denaturing solvent in protein chemistry [35], and indeed, also with PLLA, its b₀ value is relatively low at about 274. Among the solvents mentioned above, there is only one chlorinated solvent: 1,1’,2-trichlorethane, whose a₀ is anyway in the range of other chlorinated solvents, like for instance CH₂Cl₂.

3.2. On the Interaction Between PLLA-40kDa and PPA in Solution

3.2.1. Expected PLLA-40kDa and PPA Interaction in Solution

In the previous sections, we have shown that PLLA-40kDa is characterized by a ready solubility in solvents like THF, CHCl₃ and CH₂Cl₂. Furthermore, the helical fraction in any solvent studied is higher with PLLA-40kDa solutions with respect to PLLA-3D. For these reasons, in the following study on the interaction of PLLA with polyphenylacetylene (PPA), the PLLA of exclusive choice was PLLA-40kDa.

The complementarity between PLLA and PPA in terms of helical structures is represented on one side by the relative rigidity of the polyester chain in contrast with a highly dynamic PPA helix in solution [43,44]. Furthermore, PPA is characterized by a low helix-inversion barrier [42,43,44], meaning that its chains fluctuate rapidly between left- and right-handed states in a racemic solution unless dictated otherwise by a template. The template in this study is represented by the helical and intrinsically chiral PLLA-40kDa in solution.

3.2.2. Determination of the Equilibrium Association Constant in the PLLA-40kDa and PPA Interaction in CHCl₃ Solution

For the determination of the equilibrium association constant of the reaction:

[PLLA] + [PPA] → [PLLA-PPA]_complex

(6)

The Scott plot (a variation of the Benesi–Hildebrand equation was used in the condition of [PLLA] >> [PPA] [14,60]:

{[PLLA][PPA]}/A_λ = [PLLA]/E_λ + [1/(K•E_λ)]

(7)

Since in the graph, the slope = 1/E_λ while the intercept = 1/(K•E_λ), the ratio slope/intercept gives the association constant K.

With the Lambert–Beer law:

A_λ = [Complex] L•E_λ

(8)

The A_λ is absorbance at a selected wavelength, L is the path length of the UV-VIS cuvette (1 cm), and E_λ is the molar extinction coefficient at the selected wavelength. The PLLA absorption peak occurs at 209.5 nm [16], the selection of the wavelength for the study of the complex formation was made at wavelengths far away from the PLLA electronic transition, where the absorbance of the latter is negligible. Furthermore, if an increase in the baseline of the spectra was detected, then a baseline correction was introduced accordingly in the calculations.

Figure 8a shows the electronic absorption spectra of PPA in CH₂Cl₂. The stepwise addition of PLLA-40kDa causes an evident hyperchromic effect (evidenced by an arrow) in the spectral window comprised between 230 and 265 nm. For the determination of the equilibrium constant, the absorbance was selected at 250 nm and plotted according to Equation (7) in Figure 9, where it is possible to observe the perfect linearity, which is a confirmation of 1:1 stoichiometry of the PLLA-PPA complex. From the ratio slope/intercept, the equilibrium association constant for the PLLA+PPA complex was derived: K = 1103 M⁻¹ in CH₂Cl₂ at 16 °C. A similar procedure was adopted for the determination of the equilibrium association constant of the PLLA-PPA complex in THF. Figure 8b shows the electronic absorption spectra recorded after the stepwise addition of PLLA-4kDa to a solution of PPA in THF at 18 °C. Also, in this case, the hyperchromic effect was evident and suggested by the direction of the arrow. The light increase in the baseline was, of course, subtracted from the absorbances used for the determination of K in THF. As summarized in Table 3, for the determination of the association equilibrium constant of the PLLA-PPA complex in THF, the absorbance at three different wavelengths (λ = 247, 315, 400 nm) was used in the Scott plot, yielding similar K values, which were then averaged. Thus, in THF at 18 °C K = 1539 M⁻¹, at 25 °C K = 831 M⁻¹ and at 35 °C K = 560 M⁻¹ (see Table 3, where the uncertainty of the K values is reported). The association equilibrium constants of the PPA-PLLA complex were also determined in chloroform at three different temperatures and using two wavelengths: 240 and 250 nm. The results are all collected in Table 3, including also the K value of dichloromethane, which was measured only at 16 °C because of the low boiling point of this solvent.

Figure 8. Hyperchromic effect due to the stepwise addition of PLLA-40kDa to a solution of PPA (a) of 2.0 mg in 100 mL of CH₂Cl₂ at 16 °C or (b) 2.6 mg in 100 mL THF at 18 °C, in both cases in thermostated UV cuvettes. The arrow in both figures shows the hyperchromic effect due to the stepwise addition of PLLA-40kDa.

Figure 9. Example of the Scott plot according to Equation (7) due to the stepwise addition of PLLA-40kDa to a solution of PPA 2.0 mg in 100 mL CH₂Cl₂ at 16 °C. The absorbance was read at 250 nm and the excellent linearity confirms 1:1 complex stoichiometry. Since both abscissa and ordinate were multiplied by 10⁶ factor, the slope is 1.71 × 10⁻⁴ and the intercept 1.55 × 10⁻⁷, hence K_eq = 1103 M⁻¹.

Table 3. Association constants K for PLLA-PPA complex formation at different temperatures.

Solvent	T (°C)	Λ (nm)	K (M−1)
Tetrahydrofuran	18 °C	247/315/400	1539 ± 112
Tetrahydrofuran	25 °C	247/315/400	831 ± 60
Tetrahydrofuran	35 °C	247/315/400	560 ± 41
Chloroform	10 °C	240/250	1293 ± 91
Chloroform	20 °C	240/250	1080 ± 76
Chloroform	30 °C	240/250	805 ± 56
Dichloromethane	16 °C	250	1103 ± 77

To complete the thermodynamic analysis of the reaction (6), the K values determined at different temperatures either in THF or in CHCl₃ were plotted according to the Van’t Hoff equation:

Ln K = (−ΔH/R)(1/T) + (ΔS/R)

(9)

In Figure 10, the Van’t Hoff plot can be observed. From the slope, ΔH° was determined, while from the intercept ΔS°. The results of this analysis are collected in Table 4, with ΔG° calculated according to:

ΔG° = ΔH° − TΔS°

(10)

Figure 10. Van’t Hoff plot of the equilibrium constants (Ln K) for the PLLA-PPA complex against the reciprocal absolute temperature. From this graph, the enthalpy and entropy of formation in two different solvents were determined and summarized in Table 4.

Table 4. Thermodynamic parameters for PLLA-PPA complex formation in different solvents.

Solvent	ΔH° (kJ/mol)	ΔS° (J mol−1 K−1)	ΔG° (kJ/mol)
Tetrahydrofuran	−43.6	−89.17	−17.0
Chloroform	−16.9	+0.210	−16.8
Dichloromethane	n.d	n.d.	−16.8

The free energy of formation of the PLLA-PPA complex is negative and essentially the same (ΔG° ≈ −17 kJ/mol) in any solvent considered. This confirms that the complex formation is a spontaneous reaction. Table 4 also permits a breakdown of the PLLA-PPA complex formation in terms of enthalpy and entropy. In CHCl₃ (and presumably in CH₂Cl₂), the complex formation is an exothermic reaction ΔH° = −16.9 kJ/mol, without any entropic change ΔS° ≈ 0 J mol⁻¹ K⁻¹. This means that the reaction (6) in chloroform (and most probably also in CH₂Cl₂) is essentially an enthalpy-driven process. The situation is more complex in THF, where the exothermicity of reaction (6) is ΔH° = −43.6 kJ/mol, about 2.6 times that in chloroform. Such a large and negative enthalpy suggests a very strong and specific interaction occurring between PLLA and PPA in THF, and it is accompanied by a significant reduction of enthalpy ΔS° = −89.17 J mol⁻¹ K⁻¹, not seen in chlorinated solvents.

The K values relative to PLLA-PPA complex formation (Table 3) and the relative thermodynamic parameters (Table 4) suggest that in solution, a specific lock and key fit occurs between the helices of these polymers, and the phenomenon appears more facilitated in THF, although it occurs as well in chlorinated solvents.

A further insight regarding the PLLA-PPA interaction may be derived from the hyperchromic effect observed in the spectra of Figure 8. It occurs in any solvent in the 230–260 nm range, corresponding to the π → π* electronic transitions of the PPA phenyl rings.

This suggests that the environment of the PPA phenyl groups is altered because, for instance is straightened or tightened by the PLLA interaction. This is consistent with a conformation change induced by binding. Furthermore, the hyperchromic effect is extended to longer wavelengths in THF solvent (see Figure 8b). In fact, the equilibrium constant for the PLLA-PPA complex is also derivable from the absorbances at 315 and 400 nm (see Table 3), and this means that only in THF the conformation change of PPA is much more intensive, affecting also the π → π* electronic transitions of the polyenic backbone of the polymer and not only the pending phenyl rings (see Figure 2). This is fully consistent with the thermodynamic analysis and supports a chiral templating mechanism leading to a helical supramolecular PLLA-PPA adduct.

It is worth mentioning here that similar equilibrium association constants (K values) as those determined here for PLLA-PPA complex were measured in the helical complexes of PLLA-C₆₀ (K ≈ 955 M⁻¹) [15] or in the other fullerene helical complex PLLA-C₇₀ (K ≈ 1084 M⁻¹) [16]. With smaller molecules like iodine, the equilibrium constants result in smaller, e.g., PLLA-I₂ (K ≈ 25 M⁻¹), poly(methylmethacrylate)-I₂ (K ≈ 40 M⁻¹), and poly(vinylpyrrolidone)-I₂ (K ≈ 600 M⁻¹) [14].

Other host-guest complexes to be compared with the ΔG° ≈ −17 kJ/mol (Table 4), cyclodextrine-naphthenic acids show ΔG° between −16 to −28 kJ/mol [67], triphenylene-2,6,10-tricarboxamide (TTA) based supramolecular homopolymers show ΔG° ≈ −35 kJ/mol with large enthalpic effect ΔH° ≈ −50 kJ/mol but also large and negative entropy values ΔS° ≈ −50 J mol⁻¹ K⁻¹ [68]. The thermodynamic parameters of the latter system are extraordinarily similar to our values on PLLA-PPA complex in THF (Table 4), supporting our interpretation of the helical supramolecular adduct. In helical, hydrogen-bonded supramolecular self-assembled polymers, it is also possible to reach ΔG° values of −40 kJ/mol [69].

3.2.3. Determination of the Induced Cotton Effect in PLLA-PPA Complex

In Figure 8, it is possible to observe the broad band between approximately 300 and 400 nm, with a peak at about 360 nm due to the π → π* transition of the polyenic backbone of PPA either in CHCl₃ or in THF. As already discussed in the previous sections (see Figure 3 and Figure 4), the ORD of pure PLLA is plain negative, since the Cotton effect due to the electronic transition of the ester groups chromophore occurs much below 250 nm [6,16]. Thus, also in Figure 11, the blue dots show simply the plain negative curve of PLLA-40kDa in CHCl₃. However, the presence of PPA in the solution causes an induced Cotton effect, as shown by the orange dots in Figure 11, a clear indication and confirmation of the strong interaction occurring between the PLLA helices and the PPA helices in solution.

Figure 11. ORD in CHCl₃: PLLA-40kDa 423 mg/100 mL (blue dots); PLLA-40kDa 423 mg/100 mL +1.8 mg PPA (orange dots). Note the induced Cotton effect starting with a trough at about 360 nm.

The induced Cotton effect shown in Figure 11 is due to the PLLA interaction with PPA, influencing the π → π* transition of the polyenic backbone of the latter polymer. Unfortunately, in Figure 11, it is possible to see only part of the induced Cotton effect because of the limited spectral range covered by our spectropolarimeter. Anyway, it represents a further confirmation of the chiral templating mechanism leading to a helical supramolecular PLLA-PPA adduct. Such a kind of interaction between optically active polymers with guest molecules or guest polymers detectable by ORD has been known for a long time in the literature [35,36,37,38]. Furthermore, the induced Cotton effect was successfully detected also in our previous studies on PLLA chiral interaction with iodine [14], C₆₀ and C₇₀ fullerenes [15,17], and with certain dyes [16]. In the latter case, a multiple induced Cotton effect was detected in the solid state on a thin solid film of PLLA loaded with two dyes [16].

4. Conclusions

For a long time, it was disputed whether PLLA is able to keep a helical conformation in solution [5,6,7]. Recently, it was confirmed through vibrational circular dichroism that PLLA assumes a certain degree of helical conformation in solution [8,9,10]. In this work, we have examined the ORD behavior of PLLA in 13 different solvents, plus other 6 solvents from ORD literature data, in order to find the most helicogenic solvents, i.e., the solvents that are able to favor the highest degree of helical conformation of PLLA in solution. Thus, ORD data were analyzed through the Moffitt model and the Moffitt-Yang Equations (3) and (4), and Table 1 shows the solvents ordered according to the b₀ Moffitt-Yang parameter, which gives the level of helical conformation in solution. The most structuring solvents which promote a high degree of helical conformation of PLLA are ethyl acetate (b₀ = 411.1), acetonitrile (b₀ = 349.0), chlorobenzene (b₀ = 337.9), chloroform (b₀ = 328.9), and dichloromethane (b₀ = 325.5). At an intermediate level can be classified the cyclic ethers 1,4-dioxane (b₀ = 314.6) and tetrahydrofuran (b₀ = 303.9) together with toluene (b₀ = 324.7) and N,N’-dimethylformamide (b₀ = 306.9), m-cresol (b₀ = 305.7), trifluoroethanol (b₀ = 298.8) and 1,1,2-trichloroethane ((b₀ = 297.3). Acetone (b₀ = 240.4), trifluoroacetic acid (b₀ = 273.9), and sulfuric acid (b₀ = 199.6) are the least helicogenic or even denaturing solvents. An anomalous and unique behavior was observed in the case of N-methyl-2-pyrrolidin-2-one. Although it is a good PLLA solvent, it gives an anomalously low specific optical rotation at any wavelength, suggesting that it is not helicogenic at all. However, the Moffitt-Yang parameter b₀ results in extremely high values, but it is not at all corroborated by the [α]₃₈₀ value, and hence was excluded from the correlations between the b₀ and a₀ Moffitt-Yang parameters reported in Figure 6 and the general discussion.

Two PLLA grades were used in the present study, a high molecular weight PLLA-3D, designed for 3D printing, and a medium molecular weight PLLA-40kDa hexadecyl-terminated. It turned out that PLLA-40kDa is readily soluble in the solvents where it was studied, and it gives a systematically higher helical content in solution with respect to PLLA-3D in the same solvents.

Thus, PLLA-40kDa was then used as a chiral template in complex formation with poly(phenylacetylene) or PPA in solution of CH₂Cl₂, CHCl₃ and THF. In fact, PPA also maintains a helical conformation in solution, but it is racemic. In the presence of the PLLA-40kDa chiral helices, a chiral helicity induction occurs as confirmed by the extrinsic Cotton effect observed in the ORD of PLLA-PPA solutions (Figure 11). This is certainly the most definitive proof of complex formation between the two helical polymers in solution, because the extrinsic Cotton effect implies that the chiral PLLA chains with helical conformation are forcing the PPA chains into a preferred helical handedness. This can only occur by helical-helical interaction between the two polymers. Indeed, the PLLA-PPA complex formation in solution is corroborated by favorable and relevant equilibrium association constants K = 1539 ± 112 M⁻¹ at 18 °C in THF, K = 1293 ± 91 M⁻¹ at 10 °C in CHCl₃ and K = 1103 ± 77 M⁻¹ at 16 °C in CH₂Cl₂, leading to a series of highly favorable thermodynamic parameters (Table 4).

Appendix A.1

In this section, selected solvent properties and descriptors are collected.

Table A1. Properties and descriptors of the 19 solvents and PLLA studied in this paper.

	Refractive Index nD (*)	ET(30) (kcal/mol) (**)	Nile Red E(NR) (kcal/mol) (***)	Dielectric Permittivity (#)	Polarizability (Ǻ3) (*)	Hansen (†) Dispersion (MPa)1/2	Hansen (†) Polarity (MPa)1/2	Hansen (†) H- Bonding (MPa)1/2
Ethyl Acetate	1.372	38.1	54.94	6.02	8.9	15.8	5.3	7.2
Acetonitrile	1.3442	45.6	53.78	37.5	4.5	15.3	18.0	6.1
o-Dichlorobenzene	1.5514	38.0		9.93	14.3	19.2	6.3	3.3
Chlorobenzene	1.5248	36.8	53.34	5.62	12.3	19.0	4.3	2.0
Chloroform	1.4459	39.1	53.18	4.81	8.4	17.8	3.1	5.7
Dichloromethane	1.4242	40.7	53.42	9.08	6.5	18.2	6.3	6.1
Toluene	1.4969	33.9	54.81	2.38	12.3	18.0	1.4	2.0
Dioxane	1.4224	36.0	55.36	2.21	8.6	19.0	1.8	7.4
N,N’-Dimethylformamide	1.4305	43.2	52.83	36.7	7.9	17.4	13.7	11.3
Tetrahydrofuran	1.4073	37.4	56.68	8.2	7.9	16.8	5.7	8.0
1,1’,2,2’-Tetrachloroethane	1.494	39.4	53.30	8.2	12.2	18.8	5.1	5.3
Acetone	1.3588	42.2	53.99	20.7	6.5	15.5	10.4	7.0
N-Methyl-pyrrolidin-2-one	1.470	42.2		32.2	10.6	18.0	12.3	7.2
m-Cresol	1.5398	52.4	47.49	11.8	13.1	18.0	5.1	12.9
Trifluoroethanol	1.2907	59.8	48.86	26.5	5.3	15.4	8.3	16.4
1,1’,2-Trichloroethane	1.471	40.3	53.10	8.2	10.2	18.2	5.3	6.8
Dichloroacetic acid	1.466		46.50	8.2	8.9	18.2	8.1	12.2
Trifluoroacetic acid	1.2855		45.47	39.5	5.3	15.6	9.9	11.6
Sulfuric acid	1.466			110	5.5
Poly(l-Lactic acid)	1.4586	45.8 (§)	53.44 (§)	3.5		18.6	9.8	6.0

(*) from the Chemspider website: https://www.chemspider.com/ URL accessed on 9 May 2026; (**) [70]; (***) [71]; (#) see ref. [66]; (†) [72]; (§) [73].

Table A1. Properties and descriptors of the 19 solvents and PLLA studied in this paper.

	Refractive Index nD (*)	ET(30) (kcal/mol) (**)	Nile Red E(NR) (kcal/mol) (***)	Dielectric Permittivity (#)	Polarizability (Ǻ3) (*)	Hansen (†) Dispersion (MPa)1/2	Hansen (†) Polarity (MPa)1/2	Hansen (†) H- Bonding (MPa)1/2
Ethyl Acetate	1.372	38.1	54.94	6.02	8.9	15.8	5.3	7.2
Acetonitrile	1.3442	45.6	53.78	37.5	4.5	15.3	18.0	6.1
o-Dichlorobenzene	1.5514	38.0		9.93	14.3	19.2	6.3	3.3
Chlorobenzene	1.5248	36.8	53.34	5.62	12.3	19.0	4.3	2.0
Chloroform	1.4459	39.1	53.18	4.81	8.4	17.8	3.1	5.7
Dichloromethane	1.4242	40.7	53.42	9.08	6.5	18.2	6.3	6.1
Toluene	1.4969	33.9	54.81	2.38	12.3	18.0	1.4	2.0
Dioxane	1.4224	36.0	55.36	2.21	8.6	19.0	1.8	7.4
N,N’-Dimethylformamide	1.4305	43.2	52.83	36.7	7.9	17.4	13.7	11.3
Tetrahydrofuran	1.4073	37.4	56.68	8.2	7.9	16.8	5.7	8.0
1,1’,2,2’-Tetrachloroethane	1.494	39.4	53.30	8.2	12.2	18.8	5.1	5.3
Acetone	1.3588	42.2	53.99	20.7	6.5	15.5	10.4	7.0
N-Methyl-pyrrolidin-2-one	1.470	42.2		32.2	10.6	18.0	12.3	7.2
m-Cresol	1.5398	52.4	47.49	11.8	13.1	18.0	5.1	12.9
Trifluoroethanol	1.2907	59.8	48.86	26.5	5.3	15.4	8.3	16.4
1,1’,2-Trichloroethane	1.471	40.3	53.10	8.2	10.2	18.2	5.3	6.8
Dichloroacetic acid	1.466		46.50	8.2	8.9	18.2	8.1	12.2
Trifluoroacetic acid	1.2855		45.47	39.5	5.3	15.6	9.9	11.6
Sulfuric acid	1.466			110	5.5
Poly(l-Lactic acid)	1.4586	45.8 (§)	53.44 (§)	3.5		18.6	9.8	6.0

(*) from the Chemspider website: https://www.chemspider.com/ URL accessed on 9 May 2026; (**) [70]; (***) [71]; (#) see ref. [66]; (†) [72]; (§) [73].