Crown containing naphtho and anthraquinones ... - Springer Link

2282

Russian Chemical Bulletin, International Edition, Vol. 61, No. 12, pp. 2282—2294, December, 2012

Crowncontaining naphtho and anthraquinones: synthesis and complexation with alkali and alkalineearth metal cations T. P. Martyanov,a E. N. Ushakov,a V. A. Savelyev,b and L. S. Klimenkoc aInstitute

of Problems of Chemical Physics, Russian Academy of Sciences, 1 prosp. Akad. Semenova, 142432 Chernogolovka, Moscow Region, Russian Federation. Email: [email protected]; en[email protected] bN. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of the Russian Academy of Sciences, 9 prosp. Akad. Lavrentéva, 630090 Novosibirsk, Russian Federation. Email: [email protected] cYugra State University, 16 ul. Chekhova, 628012 KhantyMansiysk, Russian Federation. Email: [email protected] Novel naphtho and anthraquinones conjugated with benzo and dibenzo18crown6 ethers were obtained. Their complexation reactions with Groups Ia and IIa metal perchlorates in acetonitrile were studied by spectrophotometric titration. In most cases, the complexation involves the crown ether moiety; the stability constant of the resulting complex decreases in the following order: Ba2+ > Sr2+ > Ca2+ > Na+. For crowncontaining anthraquinone imines characterized by prototropic "imine—enamine" tautomerism, the complexation shifts the equi librium toward the imine species, which allow these compounds to be classified among a rarely occurring type of tautomeric chromoionophores. Unlike other cations, the magnesium ions preferably interact with the heteroatoms of the anthraquinone moiety (the imine N atom, the OH group, and the carbonyl O atom of the benzamido group); the logK value reaches 4.4. The chelation to the Mg2+ cations and the effect of the complexation on the tautomeric equilibrium was confirmed by quantum chemical calculations. Key words: naphthoquinone, anthraquinone, crown ether, chromoionophore, prototropic tautomerism, complexation, parameterized matrix modeling, spectrophotometric titration.

The quest for highly selective analogs of natural iono phores such as shuttling proteins in the Na+/K+ channel, enzyme cofactors, Mg2+—ATP, etc. resulted in the syn thesis of a great number of macrocyclic compounds.1 Crown ethers were among the first selective ionophores obtained.2,3 Later, they have been used to synthesize a wide range of compounds with additional (apart from ion trans fer) functions including chromophoric, luminophoric, photochromic, electrochromic/electrochemical, and other properties.4,5 Polyfunctionalized crown compounds can be employed as the basis for design of optical or electro chemical sensors capable of detecting metal cations and some organic ions, controlled molecular devices and ma chines, biometric probes, and models of biological systems. Among chromogenic crown compounds, quinone de rivatives are of particular interest. They feature an addi tional electrochromic/electrochemical function (i.e., the tendency toward the reversible redox quinonehydroquino ne transition6). For this reason, the synthesis of novel crowncontaining quinones and a quantitative study of their complexing properties are of considerable theoreti

cal and practical interest. Quantitative investigations of the ionophoric characteristics of such compounds are cur rently represented by only a few examples.7,8 Earlier,9—12 it has been demonstrated that quinones are convenient chromophores for the design of optical molecular sensors (donoracceptor chromoionophores4). In the present work, we describe the synthesis of novel crowncontaining chemosensors having one or two chromo phoric naphtho or anthraquinone moieties in their struc tures and report the results of their complexation with Group Ia and IIa metal perchlorates in acetonitrile solution. Results and Discussion Crowncontaining naphthoquinone 1 was obtained by nucleophilic substitution of the Cl atom in 2,3dichloro 1,4naphthoquinone (2) by the arylamino group of 4aminobenzo18crown6 (3) (Scheme 1).13 Heating of an ethanolic solution of the starting compound 2 and crown ether 3 in the presence of a base followed by workup

Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 12, pp. 2261—2273, December, 2012. 10665285/12/61122282 © 2012 Springer Science+Business Media, Inc.

Crowncontaining naphtho and anthraquinones

Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012

and separation of the reaction mixture by column chro matography afforded 4(2chloro1,4naphthoquinon3 yl)aminobenzo18crown6 (1) as the major reaction product (55% yield). Scheme 1

A similar procedure with dibenzocrown ethers 4a,b gave crown derivatives 5a,b containing two naphthoquinone moieties (Scheme 2). Macrocyclic 9imino derivatives of 9,10anthraquino ne were obtained photochemically as proposed earlier.14,15 This method involves irradiation of a mixture of photo chromic 1aryloxy9,10anthraquinone (6a) and ami nobenzocrown ether 3 in benzene (Scheme 3). When compound 6a is exposed to light, the aryl sub stituent migrates to the O atom in the periposition, thus

forming 9aryloxy1,10anthraquinone (7a). 1,10An thraquinone derivatives are highly reactive toward nucleo philic agents and hence the aryloxy group can easily (at ~20 C) be replaced by an arylamino group.16 The target product 4´(2benzoylamino1hydroxy9,10anthraquin on9imino)benzo18crown6 (8) was obtained in 70% yield (see Scheme 3). Crown ethers with two chromogenic groups were syn thesized using the same procedure. Photocondensation of 1aryloxy9,10anthraquinones 6 with 4,4´ and 4,5´di aminodibenzo18crown6 ethers (4a,b) gave isomeric 4´,4 and 4´,5 bis(2benzoylamino1hydroxy9,10 anthraquinon9imino)dibenzo18crown6 ethers (9a,b) in high yields and 4´,5bis(4benzoylamino1hydroxy 9,10anthraquinon9imino)dibenzo18crown6 (9c) (Scheme 4). Complexation of 1,4naphthoquinonylaminobenzocrown ethers. Complexation of crowncontaining naphthoquino nes 1 and 5a,b with alkali and alkalineearth metal cations in acetonitrile was studied by spectrophotometric titration (SPT). The concentration of metal perchlorate in solution was varied from 0 to ~0.001 mol L–1. The concentration of the ligand was kept constant during SPT. The SPT data obtained for the system 1—Ca(ClO4)2 are shown in Fig. 1. The electronic absorption spectra of naphthoquino nes 1 and 5a,b show a wide longwavelength band (LB) with a maximum at 494—496 nm. Addition of alkali and alkalineearth metal perchlorates to solutions of these compounds (C = (1—5)•10–5 mol L–1) shifts the LB to the shorter wavelengths, thus indicating complexation at the crown ether moiety.3—5 The stability constants and electronic absorption spec tra of the complexes were calculated using parameterized matrix modeling of SPT data.17 In all cases, experimental

Scheme 2

4: X = H, Y = NH2 (a); X = NH2, Y = H (b) 5: R1 = H, R2 = Nq (a); R 1 = Nq, R2 = H (b)

Nq =

2283

2284


Martýanov et al.

Scheme 3

7a

Scheme 4

4: X = H, Y = NH2 (a); X = NH2, Y = H (b) 6: R1 = NHCOPh, R2 = H (a); R1 = H, R2 = NHCOPh (b) 7: R1 = NHCOPh (a, b), H (c); R2 = H (a, b), NHCOPh (c) 9: R1 = NHCOPh (a, b), H (c); R2 = H (a, b), NHCOPh (c); R3 = H (a), Aq (b, c); R4 = Aq (a), H (b, c)

data were approximated well by an equilibrium reaction leading to a 1 : 1 complex L + Mn+

L•Mn+,

(1)

Aq =

where L is the ligand molecule, Mn+ is the metal cation, and K1:1 is the stability constant of the complex. The logK1:1 values and the spectral characteristics of the com plexes are given in Table 1.



D

0.3

2

1

0.1

500

600

/nm

Fig. 1. Electronic absorption spectra of compound 1 in MeCN (CL = 4.4•10–5 mol L–1) at different concentrations of Ca(ClO4)2 (CM = (0—1.3)•10–4 mol L–1): (1) free ligand and (2) com plex 3•Ca2+.

Table 1. Stability constants of the complexes of compounds 1 and 5a,b with alkali and alkalineearth metal cations and the spectrophotometric characteristics of these compounds and their complexes* Complex 1 1•Li+ 1•Na+ 1•K+ 1•Mg2+ 1•Ca2+ 1•Sr2+ 1•Ba2+ 5a 5a•Na+ 5a•K+ 5a•Mg2+ 5a•Ca2+ 5a•Sr2+ 5a•Ba2+ 5b 5b•Na+ 5b•K+ 5b•Mg2+ 5b•Ca2+ 5b•Sr2+ 5b•Ba2+

logK

2.16 4.78 5.06 2.36 8.16 8.07 8.92 4.78 4.60 2.74 6.18 7.01 7.18 4.82 4.62 2.80 6.26 7.11 7.29

max/nm

494 486 488 486 481 478 477 480 496 486 488 477 479 480 480 494 488 490 477 479 480 480

max•10–3 /L mol–1 cm–1

6.0 6.20 6.06 6.11 6.32 6.23 6.25 6.04 10.0 10.6 10.8 11.1 11.1 10.9 11.0 6.7 6.83 6.88 7.13 7.00 6.97 7.07

–/nm — 8 6 8 13 15 15 15 — 10 8 19 17 16 16 — 6 4 17 15 14 14

* In MeCN (water content < 0.03 vol.%) at 25  C, C L = = (1—5)•10–5 mol L–1, CM = 0—0.001 mol L–1, K = [L•Mn+]/ ([L][Mn+]), the total measurement error of K is within ±20% (the standard error for K in the modeling of the SPT data does not exceed 1%), max is the position of the LB maximum, max is the molar absorption coefficient at max,  = max(complex) – – max(dye).

2285

Benzocrowncontaining naphthoquinone 1 has a low er cationinduced hypsochromic effect  compared to a phenylazacrown analog10 but the stability constants of its complexes are substantially higher. Figure 2 allows one to compare the plots of logK vs. the metal cation radius (rM) for the complexes of crown containing naphthoquinone 1 and its bischromophoric analog 5a with alkali and alkalineearth metal cations. The stability of the complexes of compounds 1 and 5a,b with alkalineearth metal cations increases with an in crease in rM, which is typical of benzo and dibenzo18 crown6 derivatives.18 A similar trend is observed in the plot of logK vs. rM for complexes 1 with alkali metal cat ions, which agrees with the literature data for benzo18 crown6 ethers.19 The stability constants of the Ca2+, Sr2+, and Ba2+ complexes with naphthoquinone 1 are much higher than those of the complexes derived from isomeric bisnaphthoquinones 5a,b. At the same time, the logK val ues of complex 1•Mg2+ are slightly lower than those of complexes 5a,b•Mg2+. It is worth noting that the Na+/K+ selectivity is inverted when moving from naphthoquinone to bisnaphthoquinone. The appreciable differences be tween compounds 1 and 5a,b in complexing ability and selectivity are due to replacement of the single Csp3—Csp3 bond in benzocrown ether 1 by an aromatic Csp2—Csp2 bond, which reduces the electron density on two O atoms and makes the macrocycle conformationally less flexible and slightly contracted. Complexation of 9,10anthraquinon9iminobenzo crown ethers. 1Hydroxy9,10anthraquinone 9imines are known to exist as mixtures of two tautomers (imine and enamine species).21 The position of tautomeric equilibri um depends on a variety of factors such as the substituent

logK 9

1´

8 2´ 7 6 1 5 2

4 3 2

Li+ Mg2+ 0.06

Na+ Ca2+ 0.08

0.10

Sr2+ 0.12

K+ Ba2+ rM/nm

Fig. 2. Plots of logK vs. the metal cation radius rM for the com plexes of compounds 1 (1) and 5a (2) with alkali and alkaline earth metal cations in MeCN. The rM values are taken from Ref. 20.

2286


nature, the polarity of the solvent, and the phase state. Earlier,16 we have proved that 9arylimino1hydroxy 9,10anthraquinones predominantly exist in the imine form: the imine/enamine ratio in deuterated acetonitrile is ~4 : 1 (15N NMR data). The electronic absorption spectrum of crown com pound 8 in acetonitrile is shown in Fig. 3. The spectral pattern suggests the presence of two unresolved bands in the visible region. To position the maxima of these bands (max), we approximated the spectral curves with two log normal22 functions (see Fig. 3). The band with max = 435 nm was assigned to the imine tautomer of compound 8, while the less intense band with max = 545 nm was assigned to the enamine tautomer (Scheme 5). Crown compounds 9a,b have similar absorption characteristics. Scheme 5

Imine (435 nm)

Martýanov et al.

•103/L mol–1 cm–1

8 1

6

3

2

4

2

4

0 400

500

600

/nm

Fig. 3. Experimental electronic absorption spectrum of crown compound 8 in acetonitrile (1) and its approximation (2) by the sum of two functions lognormal (3, 4).

shifts the prototropic equilibrium "imine—enamine" to ward the imine tautomer. This conclusion agrees with the results of quantum chemical calculations (see next sec tion). In addition, complexation is accompanied by small hypsochromic shifts of the absorption bands due to both tautomers. The observed hypochromic and hypsochromic changes are due to the electronwithdrawing effect of the cation present in the crown cavity on the chromophoric fragment. For most of the ligand—metal salt systems studied, the SPT patterns were similar to that observed for the system 8—Ba(ClO4)2 (except for the system 9a—Mg(ClO4)2, see below). Using parameterized matrix modeling,17 we found that SPT data are always (except for titration with the magne sium salt) approximated well by an equilibrium reaction D

Enamine (545 nm)

Complexation of crowncontaining imines 8 and 9a,b with alkali and alkalineearth metal cations in acetonitrile was studied by spectrophotometric titration. The concen tration of metal perchlorate in solution was varied from 0 to ~0.01 mol L–1. The concentration of the ligand was kept constant during SPT. The SPT data obtained for the system 8—Ba(ClO4)2 are shown in Fig. 4. With an increase in the concentration of the metal salt, the absorption by the enamine tautomer of compound 8 (~540 nm) becomes substantially less intense, while the absorption band of the imine tautomer (~430 nm) in creases. Similar spectral changes have been noted earlier for 1hydroxy9,10anthraquinone 9arylimines as the sol vent polarity decreases.16 Most likely, an interaction of a metal cation with the crown ether moiety of compound 8

0.6

0.4

2

1

0.2

400

500

600

/nm

Fig. 4. Electronic absorption spectra of crown compound 8 in MeCN (CL = 1.2•10–5 mol L–1) at different concentrations of Ba(ClO4)2 (CM = (0—3)•10–5 mol L–1): (1) free ligand and (2) complex 8•Ba2+.



leading to a 1 : 1 complex. When treating the matter ri gorously, these ligand—metal cation systems contain four lightabsorbing components whose concentrations are determined by the equilibrium reactions shown in Scheme 6. Scheme 6

Here, Im and En are the imine and enamine tautomers of crowncontaining anthraquinones, respectively, and K1—K4 are the equilibrium constants. Obviously, the ratios of the concentrations [Im]/[En] and [Im•Mn+]/[En•Mn+] do not vary with the concen tration of the metal cation; so such systems can formally be described by two lightabsorbing components [L] = = [Im] + [En] and [L•Mn+] = [Im•Mn+] + [En•Mn+] and complexation in these systems, by an equilibrium of the type (1). It can easily be demonstrated that the apparent stabil ity constant K1:1 is related to the constants K1, K3, and K4 by the following equation: K1:1 = K1[(K4 + 1)/(K3 + 1)].

Since experimental data provide K3 > K4, the apparent constant K1:1 is somewhat lower than the stability con stant of the complex of imine tautomer K1. For the systems 8—Mg(ClO4)2 and 9a—Mg(ClO4)2, an acceptable error in the approximation of SPT data was achieved with two equilibrium reactions: L + Mg2+

L•Mg2+ + Mg2+

L•Mg2+,

L•(Mg2+)2.

The equilibrium constants of the complexation reac tions of compounds 8 and 9a,b with alkali and alkaline earth metal cations and the spectrophotometric charac teristics of these compounds and their complexes are giv en in Table 2. To characterize the extent of cationin duced spectral changes, we employed two parameters: the shift of the longwavelength maximum (im) in the elec tronic absorption spectrum and the relative hypochromic effect en at the absorption maximum of the enamine tautomer (see Table 2). The en values for alkali and alkalineearth metal cations amount to 0.28 and 0.51, re spectively. The maximum hypochromic effects are ob served for the complexation with Sr2+ (en = 0.49—0.51).

2287

Table 2. Equilibrium constants for the complexation reactions of compounds 8 and 9a,b with alkali and alkalineearth metal cat ions and the spectrophotometric characteristics of these com pounds and their complexesa Complex 8 8•Na+ 8 •К + 8•Mg2+ 8•Mg2+ + + Mg2+ 8•Ca2+ 8•Sr2+ 8•Ba2+ 9a 9a•Na+ 9a•Mg2+ 9a•Mg2+ + + Mg2+ 9a•Ca2+ 9a•Sr2+ 9a•Ba2+ 9b 9b•Na+ 9b•Ca2+ 9b•Sr2+ 9b•Ba2+

logKb maxim /nm

maxim•10–3 /L mol–1 cm–1

im

en c

/nm

— 4.48 4.73 4.13 1.64

435 432 432 436 432

8.9 9.70 9.77 9.06 9.73

— –3 –3 +1 –3

7.52 7.84 8.18 — 3.97 4.41 1.40

427 427 427 435 432 440 432

9.98 10.0 9.95 16.0 16.8 15.5 17.3

–8 –8 –8 — –3 +5 –3

.— 0.187 0.237 0.162 0.335 . 0.465 0.488 0.470 .— 0.278 –0.139 0.415

5.19 5.88 6.03 — 4.08 4.94 6.06 6.25

429 427 429 432 428 428 428 429

17.6 17.8 17.6 16.0 16.7 17.4 17.7 17.7

–6 –8 –6 — –4 –4 –4 –3

0.349 0.511 0.479 .— 0.259 0.425 0.497 0.459

In MeCN (water content < 0.03 vol.%) at 25 C, CL = (3—12)• mol L–1, CM = (0.3—1400)•10–5 mol L–1; maxim is the position of the LB maximum of the imine species, maxim is the molar absorption coefficient of the the imine species at maxim, im = maxim(complex) – maxim(dye). b K = [L•Mn+]/([L][Mn+]) for the complexes L : Mn+ (1 : 1), K = [L•(Mn+)2]/([L•Mn+][Mn+]) for the complexes L : Mn+ (1 : 2); the total measurement error of K is within ±20% (the standard error for K in the modeling of the SPT data does not exceed 1%). c en = [en(dye) – en(complex)]/en(dye), where en is the mo lar absorption coefficient of the enamine species at 545 (8), 558 (9a), and 548 nm (9b). a

•10–6

The plots of logK1:1 vs. rM (Fig. 5) allow convenient comparison of anthraquinone 8, bisanthraquinone 9a, and naphthoquinone 1 in complexing ability. In the case of crowncontaining anthraquinones, moving from a mono chromophore to a bischromophore changes the complex ing ability and selectivity, generally in the same way as noted above for crowncontaining naphthoquinones (see Fig. 2). The causes of these changes were discussed in the preceding section. The complexes of anthraquinone 8 are mostly less sta ble than the corresponding complexes of naphthoquinone 1. A striking exception is provided by Mg2+ complexes. Analogous conclusions can be made when comparing data for crowncontaining bisanthra and bisnaphthoquinones 9a,b and 5a,b, respectively.

2288


Martýanov et al.

logK

Scheme 7

9

1

8

2

7 3

6

1

5

2

4 3 2

Li+

Mg2+

Na+ Ca2+

Sr2+

K+ Ba2+

0.06 0.08 0.10 0.12 rM/nm Fig. 5. Plots of logK1:1 vs. the metal cation radius rM for the complexes of naphthoquinone 1 (1) and anthraquinones 8 (2) and 9a (3) with alkali and alkalineearth metal cations in MeCN.

Let us consider distinctive features in the complex ation of crowncontaining anthraquinones with Mg2+. First, according to SPT data, the systems 8—Mg2+ and 9a—Mg2+ produce two types of complexes: L•Mg2+ and L•(Mg2+)2. Second, the logK1:1 values with Mg2+ for com pounds 8 and 9a are higher by ~1.5 orders of magnitude than those of the corresponding naphthoquinones 1 and 5a, while they are noticeably lower with other cations. These facts can be explained rationally if one assumes that crown containing anthraquinones contain, apart from the crown ether moiety, an additional, more efficient site for coordi nation to Mg2+. Three heteroatoms of the chromophore moiety can serve as such a site (Scheme 7). For bischromophoric molecules 9a,b, the presence of three coordinative sites could reasonably be as sumed; however, in the concentration range studied for

Mg(ClO4)2, we detected only two equilibrium complex ation reactions. The electronic absorption spectra of compounds 8 and 9a, as well as those of their Mg2+ complexes, are shown in Figs 6 and 7, respectively. Different changes observed in the electronic absorp tion spectra upon the formation of complexes 8•Mg2+

•103/L mol–1 cm–1

•103/L mol–1 cm–1

10

2

16

8

1 12 2

6 8

3

1

4 4

2

400

500

600

/nm

Fig. 6. Electronic absorption spectra of crown compound 8 (1) and complexes 8•Mg2+ (2) and 8•(Mg2+)2 (3) in acetonitrile.

3

400

500

600

/nm

Fig. 7. Electronic absorption spectra of compound 9a (1) and complexes 9a•Mg2+ (2) and 9a•(Mg2+)2 (3) in acetonitrile.



2289

Table 3. Total energies E and bond lengths d in the tautomers of compound 8 and complex 8•Ca2+, calculated by the M062X/ 631G(d)/COSMO(MeCN) method Compound

Imine 8 Enamine 8 Imine 8•Ca2+ Enamine 8•Ca2+

E/au

–2217.25201 –2217.25120 –2417.19611 –2417.19340

d/Å C(1)O—H

C(9)N—H

C(4´)—N

C(9)—N

C(1´)O—Ca

C(2´)O—Ca

1.007 1.667 1.002 1.682

1.657 1.039 1.686 1.038

1.409 1.422 1.407 1.420

1.290 1.327 1.290 1.331

— — 2.462 2.467

— — 2.444 2.451

(en = 0.162) and 9a•Mg2+ (en = –0.139) suggest a structural difference between these complexes. Most like ly, the metal cation in complex 9a•Mg2+ interacts with the heteroatoms of both chromophoric moieties (i.e., this complex has a pseudocyclic structure). This assumption agrees with the fact that complex 9a•Mg2+ is superior to complex 8•Mg2+ in thermodynamic stability (see Table 2). The formation of binuclear complex 9a•(Mg2+)2 proba bly does not affect the pseudocyclic structure, and the magnesium cation is coordinated by the crown ether moi ety. The possibility of formation of pseudocyclic complex 9a•Mg2+ was demonstrated by quantum chemical calcu lations (see below). Quantum chemical modeling. The effect of the com plexation of crowncontaining imines on the tautomeric equilibrium imine—enamine in acetonitrile was studied using a DFT approach. We used the hybrid functional M062X,23 which is more efficient than "standard" func tionals in the description of noncovalent interactions, and the basis set 631G(d). The solvent effect was included in the model with the COSMO method.24 The calculated geometries of the tautomers of complex 8•Ca2+ are shown in Fig. 8. The total energies and bond lengths in the tau tomers of compounds 8 and 8•Ca2+ are given in Table 3. The tautomers of compound 8 are very close in energy; i.e., the calculated data suggest the coexistence of both structures in solution. The imine tautomer of compound 8 is slightly more preferable (E(enamine) – E(imine) = = 2.1 kJ mol–1), in agreement with experimental data (see Fig. 3). Upon complexation, the difference between the total energies of the enamine and imine tautomers in creases (E(enamine) – E(imine) = 5.9 kJ mol–1); i.e., the equilibrium is shifted toward the imine tautomer, which agrees with the experiment as well (see Fig. 4). The cationinduced shifts of the tautomeric equilibrium imi ne—enamine, which are observed for anthraquinones 8 and 9a,b, allow these compounds to be classified among a rarely occurring type of crowncontaining tautomeric chromoionophores.25,26 To estimate the efficiency of two different coordina tive sites in crown compound 8 with respect to Mg2+, we calculated the energy of a hypothetical transfer of the mag nesium cation from the crown ether cavity to the an

thraquinone moiety with participation of two MeCN mol ecules (Scheme 8). The geometries of structures 8A and 8B and the MeCN molecule in the gas phase were completely optimized by the DFT method with the functional M062X and the basis set 631G(d). The calculated models of the com

a Mg O N C

C(1´)

H C(2´) C(4´) C(9)

C(1)

b

C(1´)

C(2´) C(4´)

C(9)

C(1)

Fig. 8. 3D models of the tautomers of complex 8•Ca2+ calculat ed by the M062X/631G(d)/COSMO(MeCN) method.*

* Figs 8—10 are available in full color in the online version of the journal (http://www.springerlink.com/issn/15739171/current).

2290


Martýanov et al.

Scheme 8

8•Mg2+crown (8A)

8•Mg2+antr•(CN—Me)2 (8B)

plexes are shown in Fig. 9. The DFT energy for the reac tion in the gas phase (see Scheme 8) is –68.6 kJ mol–1.

Unfortunately, the COSMO method precludes reliable es timation of the solvation energy difference between the complexes of the types 8A and 8B.27 That is why we esti mated the energy of Mg2+ transfer in acetonitrile while considering only the solvation energy for two MeCN mol ecules (–53.9 kJ mol–1) and ignoring the solvation energy difference between the complexes. In this approximation, the transfer energy is –14.7 kJ mol–1, which is qualitative ly consistent with the experimentally substantiated as sumption that Mg2+ preferentially interacts with the heteroatoms of the anthraquinone moiety of compound 8. We also theoretically studied the structure of the pre sumed pseudocyclic complex 9a•Mg2+. Two structures of complex 9a•Mg2+ (pseudocyclic and open) are shown in Fig. 10. Both are optimized in the gas phase by the DFT method with the functional PBE28 and the basis set 3z. In structure closed(9a•Mg2+)•MeCN (see Fig. 10, a), the magnesium cation coordinates to a solvent molecule, three heteroatoms of one chromophoric moiety, and one hetero atom of the other. In structure open(9a•Mg2+)•2MeCN (see Fig. 10, b), the metal cation coordinates to two sol vent molecules and three heteroatoms of one chromo phoric moiety. The coordination bond lengths in these complexes are 1.95—2.22 Å. To estimate the relative stability of the pseudocyclic complex 9a•Mg2+, we considered a ring opening reaction involving a solvent molecule

a Mg O N C H

8A

b

closed(9a•Mg2+)•MeCN + MeCN open(9a•Mg2+)•2MeCN.

8B Fig. 9. 3D models of two complexes of compound 8 with Mg2+ calculated by the DFT/M062X method: (a) the Mg2+ cation in the crown ether cavity and (b) the Mg2+ cation coordinated to the heteroatoms of the anthraquinone moiety and two MeCN molecules. The DFT energies of structures 8A and 8B (see Scheme 8) in the gas phase are –2416.9330 and –2682.3439 au, respectively.

The energy of this reaction was calculated with allow ance for the energy of MeCN solvation (~–27 kJ mol–1, see above) while ignoring the solvation energy difference between the complexes closed(9a•Mg2+)•MeCN and open(9a•Mg2+)•2MeCN. In this approximation, the DFT energy for the opening of the pseudoring is ~19.7 kJ mol–1, which suggests higher stability of the pseudocyclic complex. Detailed quantum chemical analysis of struc ture 9a•Mg2+ in a solvent is very laborious. Our calcula



2291

a

Mg O N C

closed(9a•Mg2+)•MeCN

b

open(9a•Mg2+)•2MeCN Fig. 10. PBE/3zoptimized pseudocyclic (a) and open (b) structures of complex 9a•Mg2+ in the gas phase; the hydrogen atoms are omitted. The calculated DFT energies of closed(9a•Mg2+)•MeCN and open(9a•Mg2+)•2MeCN in the gas phase are –3842.6088 and –3975.2471 au, respectively.

tion only shows the fundamental possibility of formation of a pseudocyclic Mg2+ complex with crowncontaining bisanthraquinone 9a. We synthesized novel crowncontaining naphtho and anthraquinones, which act as photometric indicators for alkali and alkalineearth metal cations. 1,4Naphtho quinone derivatives bind metal cations by the crown ether moiety, which results in a hypsochromic shift of the long wavelength absorption band. The cationinduced changes in the electronic absorption spectra of crowncontaining 1hydroxy9,10anthraquinone 9imines are mainly due to the shift of the tautomeric equilibrium imineenamine to the imine tautomer. We demonstrated that these com pounds contain, apart from the crown ether moiety, an additional, more efficient site for coordination to Mg2+. This site is represented by three heteroatoms of the an thraquinone moiety. The structures of the complexes with

magnesium cations, as well as the shift of the tautomeric equilibrium upon the complexation, were confirmed by quantum chemical calculations. The results obtained can be used in the design of novel optical chemosensors from accessible derivatives of naphtho and anthraquinone and crown ethers. Experimental Acetonitrile (special purity grade, class 0, water content 7), competitive titration was used: solu tions to be mixed contained NaClO4 (C = 0.01 mol L–1) and the concentration of alkalineearth metal perchlorate was varied in a range of (0—1)•10–3 mol L–1. The stoichiometry of the complexation, the stability con stants of the complexes, and the electronic absorption spectra of the components were determined by parameterized matrix modeling of SPT data.17 In all cases, except for the systems 8—Mg(ClO4)2 and 9a—Mg(ClO4)2, the SPT data were well ap proximated with an equilibrium reaction leading to a 1 : 1 com plex. For the systems 8—Mg(ClO4)2 and 9a—Mg(ClO4)2, the acceptable error of approximation was achieved when using two equilibrium reactions (1) and (2): L•Mn+ + Mn+

L•(Mn+)2,

(2)

where L is the ligand molecule, Mn+ is the metal cation, and K1:1 and K1:2 are the equilibrium constants. Since reaction (2) occurs between charged species, the equi librium constant K1:2 depends on the ionic strength of solution. For this reason, parameterized modeling of SPT data for the systems 8—Mg(ClO4)2 and 9a—Mg(ClO4)2 was performed with allowance for the activity coefficients of charged species in solu tion according to the Debye—Hückel equation in the second approximation: log

, A = 1.824•106(T)–1.5, B = 50.29(T)–0.5,

where z is the charge of the ion, d is the conventional diameter of the ion, I is the ionic strength of solution,  is the dielectric constant of the solvent, and T is the temperature. For aceto nitrile,  = 35.85 at T = 298 K;32 therefore, A = 1.65 and B = 0.49. The conventional diameters of the magnesium cation and complex 8•Mg2+ were calculated using their quantum chemical models (see below). For Mg2+, the first coordination sphere made up of six acetonitrile molecules was considered. The cal culated d values for Mg2+•6MeCN and 8•Mg2+ are 10.1 and 13.6 Å, respectively. The conventional diameter of the quadru ply charged complex 8•(Mg2+)2 was a variable in the modeling of SPT data (the best fit was achieved for d = 19.0 Å).

2293

Parameterized modeling of the data from competitive SPT was carried out using two equilibrium reactions: L + M2+

L•M2+,

L + Na+

L•Na+,

where M2+ = Ca2+, Sr2+, or Ba2+. The stability constants KNa for the complexes L•Na+ were determined beforehand from direct titration data. Quantum chemical calculations. First, conformational screen ing for a model aminobenzocrown ether and its Mg2+ complex was carried out using molecular mechanics (MMFF94s force field33). In both cases, up to 10 000 conformers were found from which 25 conformers with relative energies of 0—3 kcal mol–1 were selected. Then these conformers were optimized (DFT, functional PBE,28 basis set 3z) with the Priroda program34 to select lowestenergy structures. A similar procedure was applied to the anthraquinone moiety and its magnesium complex. Sub sequently, the anthraquinone moiety was combined with the benzocrown ether moiety and the resulting structure 8 was opti mized (DFT, hybrid functional M062X,23 basis set 631G(d)) with the Gaussian 09 program.35 Likewise, the structures of Mg2+ complexes were calculated. The geometries of the tautomers of compound 8 and com plex 8•Ca2+ (coordination to the crown ether moiety) were cal culated as described above. The functional M062X was used because of its higher efficiency compared to "standard" func tionals in the description of noncovalent interactions. The sol vation effects were included in the model with the COSMO method.24 This method can be applied to the tautomers of com plex 8•Ca2+ because the structure of the crown ether moiety with the coordinated cation remains virtually unchanged when moving from one tautomer to the other, so the errors arising from the presence of a localized charge cancel out when calcu lating the energy difference between the tautomers. The conventional diameters of the solvate complex Mg2+•6MeCN and the complex 8•Mg2+ were calculated with the Gaussian 09 program (key word "Volume").

This work was financially supported by the Russian Foundation for Basic Research (Project No. 110300447). References 1. J. W. Steed, J. L. Atwood, Supramolecular Chemistry, John Wiley and Sons, New York, 2000, v. 1, 480 p. 2. C. J. Pedersen, J. Am. Chem. Soc., 1967, 89, 2495. 3. F. Vögtle, E. Weber, HostGuest Complex Chemistry of Macro cycles. Synthesis, Structures, Applications, Springer Verlag, Berlin, 1985, 421 p. 4. E. N. Ushakov, M. V. Alfimov, S. P. Gromov, Usp. Khim., 2008, 77, 39 [Russ. Chem. Rev. (Engl. Transl.), 2008, 77, 39]. 5. E. N. Ushakov, M. V. Alfimov, S. P. Gromov, Macrohetero cycles, 2010, 3, 189. 6. V. Ya. Fain, 9,10Antrakhinony i ikh primenenie [9,10An thraquinones and Their Applications], Photochemistry Cen ter of the Russian Academy of Sciences, Moscow, 1999, 92 pp. (in Russian).

2294


7. L. E. Echegoyen, H. Kim Yoo, V. J. Gatto, G. W. Gokel, J. Am. Chem. Soc., 1989, 111, 2440. 8. T. Saika, T. Iyoda, K. Honda, T. Shimidzu, J. Chem. Soc., Perkin Trans., 1993, 2, 1181. 9. J. P. Dix, F. Vögtle, Chem. Ber., 1980, 113, 457. 10. J. P. Dix, F. Vögtle, Chem. Ber., 1981, 114, 638. 11. G. Hollmann, F. Vögtle, Chem. Ber., 1984, 117, 1355. 12. H.G. Löhr, F. Vögtle, Acc. Chem. Res., 1985, 18, 65. 13. E. G. Lubenets, S. Z. Kusov, L. V. Ektova, Russ. Chem. Bull. (Engl. Transl.), 1994, 43, 410 [Izv. Akad. Nauk, Ser. Khim., 1994, 43, 452]. 14. L. S. Klimenko, S. Z. Kusov, V. M. Vlasov, Mendeleev Com mun., 2002, 12, 102. 15. L. S. Klimenko, S. Z. Kusov, V. M. Vlasov, E. N. Tchabue va, V. V. Boldyrev, Mendeleev Commun., 2006, 16, 224. 16. N. P. Gritsan, L. S. Klimenko, Z. V. Leonenko, I. Ya. Main agashev, V. I. Mamatyuk, V. P. Vetchinov, Tetrahedron, 1995, 51, 3061. 17. E. N. Ushakov, S. P. Gromov, O. A. Fedorova, Yu. V. Per shina, M. V. Alfimov, F. Barigilletti, L. Flamigni, V. Bal zani, J. Phys. Chem. A, 1999, 103, 11188. 18. A. Semnani, M. Shamsipur, J. Electroanal. Chem., 1991, 315, 95. 19. S. P. Gromov, A. I. Vedernikov, E. N. Ushakov, L. G. Kuz´mina, A. V. Feofanov, V. G. Avakyan, A. V. Churakov, Y. S. Alaverdyan, E. V. Malysheva, M. V. Alfimov, J. A. K. Howard, B. Eliasson, U. G. Edlund, Helv. Chim. Acta, 2002, 85, 60. 20. A. G. Ryabukhin, Izvestiya Chelyabinskogo nauchnogo tsentra [Bulletin of the Chelyabinsk Research Center], 2000, Issue 4 (in Russian). 21. A. V. Yatsenko, V. A. Tafeenko, S. G. Zhukov, S. V. Med vedev, S. I. Popov, Struct. Chem., 1997, 8, 197. 22. A. M. Mood, F. A. Graybill, D. C. Boes, Introduction to the Theory of Statistics, McGrawHill, New York, 3rd ed., 1974, p. 540.

Martýanov et al.

23. Y. Zhao, D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215. 24. A. Klamt, G. Schüürmann, J. Chem. Soc., Perkin Trans., 1993, 2, 799. 25. A. D. Dubonosov, V. I. Minkin, V. A. Bren, E. N. Shepelen ko, A. V. Tsukanov, A. G. Starikov, G. S. Borodkin, Tetra hedron, 2008, 64, 3160. 26. A. V. Tsukanov, A. D. Dubonosov, V. A. Bren, V. I. Minkin, Khim. Geterotsikl. Soedin., 2008, 1123 [Chem. Heterocycl. Compd. (Engl. Transl.), 2008, 44, 899]. 27. A. Klamt, F. Eckert, W. Arlt, Ann. Rev. Chem. Biomol. Eng., 2010, 1, 101. 28. J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865. 29. S. S. Gitis, A. I. Glaz, A. V. Ivanov, Praktikum po organi cheskoi khimii: Organicheskii sintez [The Laboratory Manual of Organic Chemistry: Organic Synthesis], Vysshaya Shkola, Moscow, 1991, p. 186 (in Russian). 30. Laboratorn technika organick chemie, Red. B. Keil, Praha, 1963 (in Czech). 31. W. H. Feigenbaum, R. H. Michel, J. Polym. Sci. A1: Polym. Chem., 1971, 9, 817. 32. J.F. Côté, D. Brouillette, J. E. Desnoyers, J.F. Rouleau, J.M. StArnaud, G. Perron, J. Solution Chem., 1996, 25, 1163. 33. T. A. Halgren, J. Comp. Chem., 1999, 20, 720. 34. D. N. Laikov, Yu. A. Ustynyuk, Russ. Chem. Bull. (Int. Ed.), 2005, 54, 820 [Izv. Akad. Nauk, Ser. Khim., 2005, 804]. 35. Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford, CT (USA), 2009.

Received July 23, 2012; in revised form November 21, 2012