Introduction

Supramolecular chemistry is a highly interdisciplinary field covering the chemical, physical, and biological features of complex chemical species that are held together and organized by means of intermolecular (noncovalent) bonding interactions. How well things fit together depends on their predisposition to do so, a matter frequently referred to as “preorganization.” Reliably predicting host–guest interactions is an important goal of supramolecular chemistry [1]. A molecular system with a preorganized and effectively functionalized recognition unit for guest molecules is ideal for host–guest interactions.

Nowadays, most molecular builders are very interested in constructing switchable molecular systems that can selectively bind different metal cations. The key to designing a successful system of this type involves the use of binding interactions that have well-defined, predictable geometric consequences. These are important aspects in the development of functional molecular devices of increasing complexity [2]. Ever since the first synthesis of crown ethers was reported by Pedersen [3], these molecules have been the focus of extensive study due to their ability to complex metal cations [4]. A large number of studies have shown that the binding properties of crown ethers are sensitive to change in conformation or effective size [5].

Azobenzenes comprise an interesting class of compounds that exhibit photoresponsive properties. Their photoisomerization properties have led to them becoming among the most common used photoresponsive molecular switches [6]. They have been incorporated into a number of supramolecular frameworks to produce ionophores for transports and photoswitchable receptors [7]. Azobenzene has the ability to undergo isomerization between the straight trans-isomer and the bent cis-isomer with light irradiation (trans ↔ cis) and thermal induction (cis → trans) (see Fig. 1), respectively. Because of their facile interconversion at appropriate wavelengths, azobenzenes have the potential to be used in optical switching and image storage devices [8–11] as well as molecular scissors [12] and as targets for coherent control in molecular electronics [13].

Fig. 1
figure 1

Schematic diagram of the trans ↔ cis isomerization of azobenzenes

Full size image

The basic requirement of a successful molecular switch is the presence of two distinct forms of the molecule that can be interconverted reversibly by means of an external stimulus, such as light, heat, pressure, magnetic or electric fields, a pH change or a chemical reaction [14]. Irradiating or heating azobenzene-containing materials induces reversible isomerization between the two isomers, making azobenzenes switchable molecules. The isomerization of azobenzenes is accompanied by significant changes in the absorption spectra and structures of the molecules. These changes can alter properties of their surrounding environment by switching them “on” or “off.”

The azobenzene moiety incorporated into the crowns (see Fig. 2) is used to change the size of the crowns and hence to modify the complexing properties of the molecules [15, 16]. The combination of a crown ether with an azobenzene moiety enables us to control ionic conductivity by light irradiation or thermal induction.

Fig. 2
figure 2

Schematic diagram of the trans ↔ cis isomerization of crown ethers with an incorporated azobenzene moiety

Full size image

The azobenzene-type crown ethers (hereafter referred to as “ACEs”) Cr(O7) and Cr(O8), in which the 4 and 4′ positions of azobenzene are linked by a polyoxyethylene chain, were synthesized and studied by Seiji Shinkai and co-workers [17]. Cr(O7) and Cr(O8) have azobenzene as an antenna and the crown ether as a functional group, and change their chemical and physical functions in response to photoirradiation or changes in temperature. Similarly, azobenzene derivatives have been utilized as light-driven or temperature-driven triggers to control the functions of metal ligands.

Computational methods are a promising way to calculate the structures and properties of complexes, such as their binding energies and absorbance spectra. In the work presented here, a family of ACEs [Cr(O6), Cr(O7) and Cr(O8)] with rings of different sizes containing the azobenzene moiety incorporated into the crown were studied theoretically.

Theory and methods of calculation

In the framework of the density functional theory (DFT) approach, the B3LYP hybrid functional [18, 19] is one of the most preferred methods, as it has proven its ability to reproduce various molecular properties, including structural parameters and vibrational spectra. The combined use of the B3LYP functional and the standard split valence basis set 6-31 G(d) has been previously shown to provide an excellent compromise between the accuracy of the results and computational efficiency for large and medium-sized molecules [20–26]. Ground-state electronic structure calculations of all complexes were performed using density functional theory (DFT) methods as implemented using the Gaussian 03 software package [27]. The functional that was used throughout this study is B3LYP, consisting of a hybrid exchange functional, as defined by Becke’s three-parameter equation, and the Lee–Yang–Par correlation functional [18, 19]. The ground-state geometries were obtained in the gas phase by full geometry optimization, and the optimum structures, located as stationary points on the potential energy surfaces, were verified by the absence of imaginary frequencies. The standard 6–31 G(d) and LANL2DZ basis sets were found to be suitable for most ligands.

Time-dependent density functional theory (TDDFT) can model highly complex molecules like azobenzenes accurately, efficiently, and cost-effectively. In this study, TDDFT was used to model the absorption spectra of several azobenzene derivatives. The results show a reasonably good association between the theoretical and experimental values for the absorbance spectra of the azobenzenes. A natural bond orbital (NBO) population analysis was performed with the NBO 3.1 program as implemented in Gaussian [28–31]. NBO analysis represents a unique and powerful approach to evaluating the origins of intermolecular interactions from a computational standpoint.

The binding energies, binding enthalpies, and Gibbs free energies in the gas phase for the complexes were calculated for the reaction

$$ cis - {\text{ACE}}{{\text{s}}_{\text{free}}} + {\text{M}}_{\text{free}}^{{ + }} \to {{(}cis - {\text{ACEs/}}{{\text{M}}^{{ + }}})_{\text{complex}}}. $$
(1)

For this system, the binding energy ΔE can be expressed as follows:

$$ \Delta E = E{(cis - {\text{ACEs/}}{{\text{M}}^{ + }})_{\text{complex}}} - [E(cis - {\text{ACE}}{{\text{s}}_{\text{free}}}) + E({\text{M}}_{\text{free}}^{ + })]. $$
(2)

Results and discussion

Optimized ground-state geometry

The structures of molecules play an especially important role in determining their chemical properties. The optimized stability structures for both the trans and cis forms of ACEs [Cr(O6), Cr(O7) and Cr(O8)] were obtained at the B3LYP/6-31 G(d) level in the gas phase at 298 K, while unsubstituted trans- and cis-azobenzene were studied as reference compounds at the same level. The results of the analysis of all of the target molecules described above are depicted in Table 1, and their ground-state structures are presented in Fig. 3.

Table 1 Comparison of the calculated parameters of the trans and cis isomers of ligands optimized at the B3LYP/6-31 G(d) level
Full size table
Fig. 3
figure 3

The optimized structures of the trans and cis isomers of azobenzene and azobenzene crown ethers (ACEs) at the B3LYP/6-31 G(d) level

Full size image

The trans isomer of azobenzene is about 15.1 kcal mol−1 or 0.65 eV lower in energy than that of the cis isomer. This is only slightly higher than the experimental value of 0.6 eV [32]. The DFT results are very similar to some of the previous theoretical predictions [33–39]. The calculated results indicate that the phenyl rings of trans-azobenzene are 50.2° out of plane compared to those of the cis isomer, and the distance between the 4 and 4′ positions decreases from 9.079 Å to 6.562 Å for trans- and cis-azobenzene, respectively.

The energies of the cis ACEs are 16.3 kcal mol−1 (cis-Cr(O6)), 20.1 kcal mol−1 (cis-Cr(O7)), and 15.7 (cis-Cr(O8)) kcal mol−1 higher than those of their corresponding trans isomers, respectively. The optimized structures of the trans isomers of Cr(O6), Cr(O7), and Cr(O8) are shown in Fig. 3, and the calculated parameters for them are listed in Table 1.The polyoxyethylene (−CH2–O–CH2–) n (n = 1, 2, 3) chains between the two aromatic rings are almost linearly extended. The distances between the 4 and 4′ positions of azobenzene of the three trans isomers are 8.945, 9.072 and 9.034 Å, respectively, which are all smaller than those of the unsubstituted trans-azobenzene (9.079 Å). The angle ∠NNCC for trans-Cr(O6), trans-Cr(O7), and trans-Cr(O8) are 3.2°, 1.4°, and 8.4°, respectively. These results indicate that the phenyl rings of the trans-azobenzene unit in the ACEs are out of plane compared to those of the unsubstituted azobenzene, and there must be some steric restriction in play during the trans ↔ cis isomerization. The methylene chain of trans-Cr(O8) undergoes a small amount of folding, and trans-Cr(O6) shows the most restricted structure (see from Fig. 3). The trans isomers of the ACEs show poor preorganization because of the long loops in their structures; the isomers lack any affinity for metal cations according to the rules of supermolecule preorganization [40]. With respect to the cis ACEs, there is a crown loop in each of the target molecules. The preorganization of the cis forms of ACEs is enhanced, allowing them to coordinate with metal cations; they thus present an “on” state, while the trans forms of the ACEs are in an “off” state in terms of coordinating with metal cations. Cis forms of these crown ethers show affinity for metal cations. The sizes of the loops in the cis-type ACEs follows the order: cis-Cr(O6) < cis-Cr(O7) < cis-Cr(O8). The arrangement of atoms in the cis isomers is more relaxed than that in the trans isomers. In addition, the flexibility of the crown-like loop increases as the number of –CH2–O–CH2– units between the 4 and 4′ positions of azobenzene increases. These results show that in ACEs with a polyoxyethylene chain, the crown-like loop appears in cis ACEs and disappears in trans ACEs, causing an “all-or-nothing” change in the ion-binding ability. The molecules of ACEs in their cis and trans isomer forms act as “switched-on” and “switched-off” crown ethers, respectively.

Optimized geometries of the complexes

The optimized structures of the cis ACE/M+ complexes [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)]/Li+, Na+, K+ and Rb+] are given in Fig. 4, whereas the most important parameters for these complexes, which were optimized by performing DFT at the B3LYP/6-31 G(d) and LANL2DZ level, are given in Table 2.

Fig. 4
figure 4

The optimized structures of the cis isomers of azobenzene crown ethers (cis ACEs) complexed with Li+, Na+, K+ and Rb+ metal cations, obtained at the B3LYP/6-31 G(d) and LANL2DZ level of theory

Full size image
Table 2 Selected parameters for the ACE/M+ (Li+, Na+, K+ and Rb+) complexes optimized at the B3LYP/6-31 G(d) and LANL2DZ level (distances in Å, dihedral angles in degrees °)
Full size table

Upon inspecting Figs. 3 and 4 and Tables 1 and 2, it is clear that the distances between the 4 and 4′ positions of the azobenzene crown ether loops all change greatly when the free ligand cis ACE [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)] coordinates with the alkali cations (Li+, Na+, K+ and Rb+). It can be assumed that the decreased distances d C–C are due to the inductive effect arising from the O…M+ interactions. The smaller the number of methylene groups between the 4 and 4′ positions of azobenzene, the stronger the restrictions on the crown-like ring. In addition, not all of the oxygen atoms can contribute to the formation of a crown-like ring in the cis isomers. The oxygen atoms in the crown loops do not all interact with M+( Li+, Na+, K+ and Rb+) because they are too far away.

When the crown ether loop of cis-Cr(O6) coordinates with an alkali-metal cation, the structural features of the dihedral angle ∠CNNC of the complex change significantly and show different properties to those of the metal-free cis-Cr(O6) (11.1°). Upon inspecting Fig. 4 and Table 2, it is clear that in the complexes cis-Cr(O6)/Li+ and cis-Cr(O6)/Na+, the interatomic distances between Li+ and (O1–O6) are 2.155, 2.096, 2.343, 2.078, 2.094, and 4.965 Å; those between Na+ and (O1–O6) are 2.405, 2.431, 2.486, 2.387, 2.457, and 5.115 Å, respectively. It is clear that r 6 is 4.965 Å for cis-Cr(O6)/Li+ and 5.115 Å for cis-Cr(O6)/Na+. The bond length r 6 is much larger than the others in each of the complexes. The oxygen O6 in the crown loop of cis-Cr(O6) shows only weak interactions with Li+ and Na+ because it is too far away from them. The optimized structure of cis-Cr(O6) with and without the cations K+ and Rb+ presents only small changes. This result can be attributed to the small size of the crown-like cavity but the big cation diameters of K+ and Rb+.

Turning our attention to the structures of the complexes cis-Cr(O7)/M+ (Li+, Na+, K+ and Rb+), a polyoxyethylene loop is formed that is analogous to 15-crown-5 [40–42]. Not all of the donor oxygen atoms in cis-Cr(O7) interact with metal cations. In the complexes, the average coordination bond lengths of the metal cations are 2.111, 2.424, 2.783, and 3.050 Å, respectively. The bond lengths for cis-CrO7/Li+ and Na+ are smaller than those for cis-Cr(O7)/K+ and Rb+; in other words, there are stronger metal–oxygen interactions between the ligand cis-Cr(O7) and Li+ and Na+ than K+ and Rb+. Li+ gives a shorter bond length with the donor O than Na+ does in these complexes. However, the structures of cis-Cr(O7)/M+ (Li+ and Na+) from Fig. 4 indicate that Na+ improves the planarity of the oxygens compared to Li+. Therefore, Na+ fits with the crown-like ring better than Li+. Cis-Cr(O7) cannot bind well with the large alkali metal cation Rb+, as can be seen from the structures in Fig. 4.

Based on the optimized structures of the complexes formed by the alkali cations Li+, Na+, K+ and Rb+ and the ligand cis-Cr(O8), a 18-crown-6 crown-like ring is produced when cis-Cr(O8) coordinates with metal cations. There are six donor oxygens that are mainly involved in the O…M+ interactions. Cis-Cr(O8) can bind with both small and large alkali-metal cations, as can be seen from the structures shown in Fig. 4. Li+ is too small to coordinate with all six oxygen atoms in the crown-like ring of cis-Cr(O8). The coordination bond lengths shown in Table 2 for the complex cis-Cr(O8)/Li+ are 3.290, 2.009, 2.055, 2.066, 2.016 and 3.192 Å, respectively. It is clear that r 1 and r 6 are all much larger than the other bond lengths. The optimized structure shown in Fig. 4 indicates that Li+ is drawn to one side of the crown-like ring. The bond lengths indicate that the best match for the crown-like loop in the ligand cis-Cr(O8) is Na+ according to the lock-and-key complementarity rule [43]. However, crown-6 ethers are known to prefer K+ according to experimental results [44–47]. The most plausible reason for this difference between experiment and theory is that the calculations do not include the effect of the solvent, and Na+ is even more strongly solvated than K+ [48]; the calculations were performed for isolated molecules in the gas phase, but the experiments were done in aqueous solution.

Natural bond orbital analysis

For each donor NBO (i) and acceptor NBO (j), the stabilization energy (E 2) associated with i → j delocalization is explicitly estimated using the following equation [49–52]:

$$ {E_2} = \Delta {E_{{ij}}} = {q_i}\frac{{{F^2}\left( {i,j} \right)}}{{{\varepsilon_j} - {\varepsilon_i}}}, $$
(3)

where q i is the ith donor orbital occupancy, ε i and ε j , are the diagonal elements (orbital energies), and F (i, j) are the off-diagonal elements, respectively, associated with the NBO Fock matrix.

The results of second-order perturbation theory analysis of the Fock matrix for cis-ACE/M+ [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)/Li+, Na+, K+ and Rb+], obtained by NBO analysis, are summarized in Table 3. The interaction energies E 2 of the host–guest molecules cis-ACE/M+ are mainly dependent on the lone-pair electrons of O atoms of the crown ether and the LP* orbitals of the alkali-metal cation (Li+, Na+, K+ and Rb+); the N atoms in the azobenzene part do not appear to be as important.

Table 3 Results of second-order perturbation theory analysis of the Fock matrix for cis-ACE/M+ [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)/Li+, Na+, K+ and Rb+] within the NBO basis
Full size table

For the complexes cis-Cr(O6)/M+, the strong donor–acceptor interactions for cis-Cr(O6)/Li+ between the lone-pair electrons of the electron-donating oxygens O1, O2, O3, O4, and O5 and the LP* orbital of Li+ have stabilization energies of 3.05, 4.76, 4.47, 3.81, and 4.41 kcal mol−1, respectively, which are much bigger than the corresponding energies E 2 for the complex cis-Cr(O6)/Na+ (1.28, 3.72, 3.14, 2.92, and 3.32 kcal mol−1). Obviously, one of the electron-donating oxygens, O6, in the ligand cis-Cr(O6) is not considered to interact with Li+ and Na+, and the E 2 data show a poor distribution. However, the interaction stabilization energy E 2 between an electron-donating oxygen and K+ or Rb+ is weaker than that for cis-Cr(O6)/M+ (Li+ and Na+). Also, six electron-donating oxygens in cis-Cr(O6) all interact with the cations K+ and Rb+. This result can be attributed to the small size of the crown-like cavity and the big cation diameters of K+ and Rb+.

In cis-Cr(O7)/M+ (Li+, Na+, K+ and Rb+) complexes, the stronger donor–acceptor interactions mainly derive from the lone-pair electrons of the five electron-donating oxygens (O1, O2, O3, O4, O5) and the LP* orbital of the alkali-metal cation M+(Li+, Na+, K+ or Rb+), and the interaction phenomenon is analogous to 15-crown-5/M+. The stabilization energies E 2 for the complexes cis-Cr(O7)/Li+ and Na+ are larger than those of the complexes cis-Cr(O7)/K+ and Rb+. For the complex cis-Cr(O7)/Li+, the data distribution for the stabilization energy E 2 of cis-Cr(O7)/Na+ is better than that of cis-Cr(O7)/Li+.

In the cis-Cr(O8)/M+ complexes, the strongest donor–acceptor interactions mainly come from the lone-pair electrons of the six electron-donating oxygens (O1, O2, O3, O4, O5, O6) and the LP* orbital of the alkali-metal cation M+(Li+, Na+, K+ or Rb+), and the interaction phenomenon is analogous to 18-crown-6/M+. The stabilization energy E 2 for O6…Li+ in the cis-Cr(O8)/ Li+ complex is 2.33 kcal mol−1, which is much smaller than those of the other five oxygens (O1: 3.10, O2: 3.54, O3: 3.29, O4: 3.30, O5: 3.51 kcal mol−1). The data distributions of the stabilization energies E 2 for the complexes cis-Cr(O8)/K+ and cis-Cr(O8)/Rb+ are better than those of the complexes cis-Cr(O8)/Li+ and cis-Cr(O8)/Na+. In the complex cis-Cr(O8)/K+, the strong donor–acceptor interactions between the lone-pair electrons of the electron-donating oxygens O1–O6 and the LP* orbital of Li+ have stabilization energies of 2.95, 2.38, 2.69, 2.72, 2.51 and 2.99 kcal mol−1, respectively, which are much bigger than the corresponding stabilization energies E 2 of the complex cis-Cr(O8)/Rb+ (1.62, 1.32, 1.51, 1.54, 1.34 and 1.76 kcal mol−1).

Binding energies and stabilities

The calculated binding energies (ΔE b), enthalpies (ΔH b) and Gibbs free energies (ΔG b) (298 K) of the ACE/M+ complexes [cis-Cr(O6), cis-Cr(O7), and cis-Cr(O8)/Li+, Na+, K+ and Rb+], based on reaction (1) at the B3LYP/6-31 G(d) and LANL2DZ level in the gas phase are listed in Table 4. When performing such a study, it is important to consider the large basis set superposition error (BSSE), which in most cases leads to overestimated interaction energies [53, 54]. One of the most commonly used methods of correcting for the BSSE is the counterpoise (CP) method [55]. Thus, the binding energies were corrected for the undesirable effects of the BSSE using the CP method at the B3LYP/6-31 G (d) level with relaxed fragment geometries.

Table 4 Calculated binding energies ΔE b (kcal mol−1), binding enthalpies ΔH b (kcal mol−1), and Gibbs free energies ΔG b (kcal mol−1) in the gas phase for the complexes at 298 K
Full size table

Table 4 shows that the gas-phase binding energies (ΔE b), binding enthalpies (ΔH b) and Gibbs free energies (ΔG b) at 298 K decrease for the three different free ligands cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8) as the size of the alkali cation increases, in other words: \( \Delta {E_{{{\text{ACEs/L}}{{\text{i}}^{ + }}}}} > \Delta {E_{{{\text{ACEs/N}}{{\text{a}}^{ + }}}}} > \Delta {E_{{{\text{ACEs/}}{{\text{K}}^{ + }}}}} > \Delta {E_{{{\text{ACEs/R}}{{\text{b}}^{ + }}}}}. \)

For cis-Cr(O6)/M+ (Li+, Na+, K+ and Rb+), the crown-like cavity ring of cis-Cr(O6) must undergo considerable folding/twisting to bring the binding sites in close proximity to the small cations Li+ and Na+. These distortions enhance the host–guest intramolecular interactions. Although the backbone of the complex suffers much distortion and displays poor structural symmetry, the calculations are performed for isolated molecules in the gas phase (i.e., they do not include the intramolecular interactions of the studied complexes); therefore, the thermal energies of cis-Cr(O6)/Li+ and cis-Cr(O6)/Na+ are larger than those of cis-Cr(O6)/K+ and cis-Cr(O6)/Rb+. However, because the metal cations K+ and Rb+ are large but the crown-like cavity is small, the thermal energies of cis-Cr(O6)/K+ and cis-Cr(O6)/Rb+ are also small. For cis-Cr(O7)/M+ and cis-Cr(O8)/M+ (Li+, Na+, K+ and Rb+), the complexes suffer much distortion, display poor structural symmetry and so exhibit the biggest thermal energies for the cation Li+. Thus, the relationship between the cavity size of the crown ether and the cation diameter plays an important role in determining the thermal energies of complexes during coordination.

For cis-Cr(O6)/M+, cis-Cr(O7)/M+ and cis-Cr(O8)/M+ (Li+, Na+, K+ and Rb+), the different alkali cations show different trends. If we consider Na+, the thermal energy shows the relation cis-Cr(O6)/Na+ < cis-Cr(O6)/Na+ < cis-Cr(O6)/Na+, while the binding energies of the cis ACEs with Na+ [cis-Cr(O6) < cis-Cr(O7) < cis-Cr(O8)] indicate that the affinities of the cis ACEs for Na+ increase as the number of –CH2–O–CH2– units increases, enlarging the crown-like loop. As the loop enlarges, the rigidity of the crown ether is reduced, so it becomes easier for the ACE to bind with metal cations.

Absorption spectra

The absorption spectra of the cis isomers of azobenzene and ACEs [cis-Cr(O6), cis-Cr(O7 ) and cis-Cr(O8)] were investigated by time-dependent density functional theory (TDDFT) with the 6-31 G(d) basis set. The calculated excitation energies (E g), wavelengths of peak absorption (λ abs) and the oscillator strengths (f) of all compounds in their optimized ground-state geometries are summarized in Table 5.

Table 5 Electronic transition data obtained by TDDFT for a family of azobenzene-type crown ethers [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)]
Full size table

The absorption spectrum of cis-azobenzene shows two distinct bands: a strong π → π*,(S0 → S2) absorption band peaking at about 270 nm and a much weaker n → π*(S0 → S1) band with a peak at around 470 nm. The results are in a good agreement with some of the previous studies [56, 57].

Obviously, the values of the parameters of the cis isomers of the ACEs are all different from those of cis-azobenzene. The spectra of the cis ACEs [cis-Cr(O6), cis-Cr(O7) and cis-Cr(O8)] contain broad π → π* (S0 → S2) absorption bands with a characteristic peak at 310–340 nm in the near-UV region, and their oscillator strengths are much more intense. Weaker bands in the visible region (peak wavelengths: 480–490 nm) and lower oscillator strengths due to the n → π* (S0 → S1) transitions are also observed. The spectra of the cis ACEs present significant redshifts in comparison to the spectrum of unsubstituted cis-azobenzene. This result indicates that the size of the crown has a distinct influence on the absorption spectra of the cis ACEs.

In ref. [15], the experimental results indicated that both Cr(O7) and Cr(O8) give high yields of the cis isomer at about 360 nm, which is a little different from the peak wavelengths obtained in our calculations. This difference between the theoretical calculations and the experimental results arises because the calculations performed in this paper relate to the gas phase at 303 K, while the experiments were performed in liquids at 298 K. However, this theoretical study is still useful for predicting reactions and gauging trends.

Conclusions and perspectives

The ground-state electronic structures of azobenzene crown ethers [ACEs: Cr(O6), Cr(O7) and Cr(O8)] and complexes of their cis isomers with the alkali-metal cations Li+, Na+, K+ and Rb+ were obtained by DFT methods at the B3LYP/6-31 G(d) level and LANL2DZ. The significant structural differences between the optimized trans and cis isomers of the ACEs indicate that the preorganization of the trans ACEs is poor and in an “off” state, while it is enhanced for the cis isomers and in an “on” state in relation to coordinating with alkali metal cations. These “molecular machines” can therefore be used as "on/off” switches as they can switch between different molecular structures and parameters. The cis isomers showed spherical recognition patterns in the binding of alkali-metal cations. In NBO analysis, the main intermolecular charge-transfer interactions were between the LP* orbitals of the metal cations and the lone-pair electrons of the electron-donating O atoms of the cis ACEs, but not all of the donor oxygen atoms in the cis ACEs interact with metal cations. The interaction pattern of cis-CrO7 with metal cations (M+) is analogous to 15-crown-5/M+, while that for cis-Cr(O8)/M+ is analogous to 18-crown-6/M+. A time-dependent density functional theory (TDDFT) study of the cis ACEs afforded their absorption spectral parameters. The results of the TDDFT study indicate that the cis isomers have broad π → π* (S0 → S2) absorption bands but weaker n → π* (S0 → S1) bands, and good agreement between the theoretical and experimental values was seen.