Solubility Measurements and Prediction of CoenzymeQ10 Solubility in Different Solvent Systems

Yin Zhao • Yan-Hong Sun • Zhi-Yong Li • Chuang Xie •

Ying Bao • Zhi-Jian Chen • Jun-Bo Gong • Qiu-Xiang Yin •

Wei Chen • Cui Zhang

Received: 17 March 2012 / Accepted: 15 September 2012� Springer Science+Business Media New York 2013

Abstract The solubility of coenzyme Q10 in ethyl acetate, n-hexane, 1-butanol, 1-pro-

panol, 2-propanol and ethanol in the temperature range 270.15–320.15 K, under atmo-

spheric pressure, was measured by a gravimetric method and compared with the data

predicted using the conductor like screening model for realistic solvation (COSMO-RS)

method. The results show that the solubilities of coenzyme Q10 in the above solvents

increase with temperature. The temperature dependences of predicted solubilities were

consistent with the experimental data. The experimental data were correlated with the

Apelblat equation. At the same temperature, the order of increasing solubility is ethyl

acetate [ n-hexane [ 1-butanol [ 1-propanol [ 2-propanol [ ethanol.

Keywords Coenzyme Q10 � Solubility � COSMO-RS � Gravimetric method �Apelblat equation

1 Introduction

Coenzyme Q10 ([(2E,6E,10E,14E,18E,22E,26E,30E,34E)-3,7,11,15,19,23,27,31,35,

39-decamethyltetraconta-2,6,10,14,18,22,26,30,34,38-decaenyl]-5,6-dimethoxy-3-methyl-

cyclohexa-2,5-diene-1,4-dione), is an endogenous lipid-soluble, vitamin-like benzoquinone

compound which is found in most eukaryotic cells, especially in the mitochondria. It is a

component of the electron transport chain and participates in aerobic respiration, gener-

ating energy in the form of ATP [1]. It also acts as a powerful antioxidant which scavenges

free radicals, prevents the initiation and propagation of lipid peroxidation in cellular

Y. Zhao � Y.-H. Sun � Z.-Y. Li � C. Xie � Y. Bao � Z.-J. Chen � J.-B. Gong � Q.-X. Yin �W. Chen (&)Department of Chemical Engineering, School of Chemical Engineering and Technology,Tianjin University, Tianjin 300072, Chinae-mail: chenwei@tju.edu.cn

C. ZhangInstitute of New Catalytic Materials Science, Key Laboratory of Advanced Energy MaterialsChemistry (MOE), College of Chemistry, Nankai University, Tianjin 300071, China

123

J Solution ChemDOI 10.1007/s10953-013-9998-5

biomembranes, and helps the regeneration of a-tocophero [2]. The molecular structure of

coenzyme Q10 is shown in Fig. 1. Knowledge of the solubility of coenzyme Q10 is

essential to optimize its production since it directly affects the yields and thus controls the

cost of its production [3]. In this work, the solubility of coenzyme Q10 in different solvents

at various temperatures under atmospheric pressure was experimentally determined using a

gravimetric method [4] and correlated with the Apelblat equation.

Over recent years, the conductor like screening model for realistic solvation (COSMO-

RS) method of Klamt [5] was developed as a novel approach to predict equilibrium

thermodynamic properties. In the present study, the COSMO-RS calculations were carried

out to evaluate the solubility of coenzyme Q10 in several solvents.

2 Experimental

2.1 Materials

Coenzyme Q10 was supplied by Shen Zhou Biology and Technology Co., Ltd., and

purified by recrystallization. The product’s mass fraction determined with an Agilent 1200

HPLC was found to be higher than 99.0 %. Ethyl acetate, n-hexane, 1-butanol, 1-propanol,

2-propanol and ethanol were purchased from Tianjin Kewei Chemical Co. in China. All of

the solvents were of analytical reagent grade and used for all experiments without further

purification.

2.2 Solubility Measurements

The solubilities of Q10 in several solvents at (270.15, 280.15 and 290.15) K were measured

by UV spectrophotometric analysis (Hitachi UV-3010). An excess of the solid coenzyme

Q10 was added to the pure solvent and then stirred in a jacketed glass vessel (50 mL). The

temperature was maintained constant within 0.02 K by circulating in a silicone oil bath

with a thermoelectric controller (CF41, Julabo, Germany). A thermometer was inserted

into the inner chamber of the vessel to confirm the temperature of the solution. The

solution was stirred using a magnetic stirrer bar for about 12 h to reach equilibrium and

then allowed to settle for 4 h. The clear solution was extracted with a 10.0 mL glass

syringe and filtered through a 0.2 lm PTFE filter. The mole fractions of Q10 in the

solutions were determined from absorbance versus concentration calibration curves

derived from the measured absorbance of solutions of known concentrations. The

O

O

O

O

H3C

H3C

CH3 CH3

H2CH2C C

HC

10

Fig. 1 The Molecular structure of coenzyme Q10

J Solution Chem

123

estimated uncertainty of the solubility values, based on error analysis and repeated

observations, was within 2.8 %.

The corresponding solubilities of Q10 at (300.15, 310.15 and 320.15) K were also

determined by a gravimetric method. The clear solution was obtained as described above

and injected rapidly into a pre-weighed petri dish (m0). The petri dish with clear solution

was quickly weighted (m1) and placed in the blast-drying oven (101 A, Shanghai Sheng

Xin Scientific Instrument Co. Ltd., China) at 303.15 K for 24 h. The petri dish was

reweighed several times until the weight of the petri dish (m2) ceased to change. Every

experiment was repeated three times. The mass of the solute and petri dish was measured

using an analytical balance (Metler Toledo AL104. Shanghai, China) with an accuracy of

±0.0001 g. The mole fraction of coenzyme Q10 in the sample solution were calculated by

Eq. 1,

x ¼ ðm2 � m0Þ=M1

ðm2 � m0Þ=M1 þ ðm1 � m2Þ=M2

ð1Þ

in which M1 and M2 are the molar mass of coenzyme Q10 and the solvent, respectively.

The solubilities of Q10 in solvents reported in Table 1 represent averages of three mea-

surements with a reproducibility better than 97 %. However, the uncertainty of the solu-

bility in ethyl acetate or n-hexane at 320.15 K was almost 20 % so those data were

excluded from the correlation described below. It was considered that coenzyme Q10

Table 1 Predicted and experimental mole fraction solubility (xpre and xexpt) of coenzyme Q10 in thedifferent solvents

T/K xpre xexpt T/K xpre xexpt

Ethyl acetate n-Hexane

270.15 0.0003 0.0006 270.15 0.0001 0.0009

280.15 0.0019 0.0033 280.15 0.0005 0.0044

290.15 0.0099 0.0178 290.15 0.0027 0.0207

300.15 0.0463 0.0900 300.15 0.0152 0.0931

310.15 0.1952 0.4309 310.15 0.1031 0.3993

320.15a 0.7492 0.6471 320.15a 0.7394 0.5503

1-Butanol 2-Propanol

270.15 0.00004 1.0E-05 270.15 0.00004 0.7E-05

280.15 0.00023 1.7E-05 280.15 0.00025 0.8E-05

290.15 0.0014 0.0001 290.15 0.0014 6.2E-05

300.15 0.0076 0.0011 300.15 0.0076 0.0005

310.15 0.0424 0.0107 310.15 0.0430 0.0029

320.15 0.6218 0.0934 320.15 0.6439 0.0165

1-Propanol Ethanol

270.15 0.00004 2.2E-05 270.15 0.00002 1.0E-05

280.15 0.00024 0.0002 280.15 0.00013 1.1E-05

290.15 0.0014 0.0005 290.15 0.00073 1.6E-05

300.15 0.0072 0.0014 300.15 0.00388 6.1E-05

310.15 0.0410 0.0039 310.15 0.02173 0.0002

320.15 0.6433 0.0110 320.15 0.62001 0.0014

a These data with a large uncertainty of 20 % were excluded from correlation with the Apelblat equation

J Solution Chem

123

formed a mixture of melt and solution in a good solvent at the temperature nearly at its

melting point of 322.35 K (Fig. 2).

For crystal structure confirmation, X-ray powder diffraction analysis were carried out

using a Rigaku-2500 Diffractometer using Cu Ka radiation in the step-scan mode

(2h = 0.02� per step). The crystals obtained from each solvent showed similar XRD patterns

to the original sample indicating that no polymorphs crystallized during experimental mea-

surement of solubilities. The solid obtained from each solvent showed the same XRD pattern

as the original crystals (Fig. 3), while no solvate was found from the DSC data (Fig. 2).

2.3 Cosmo-RS Prediction

COSMO-RS is a surface interaction model that considers the molecular charge densities

that are obtained by molecular quantum chemical calculations. A key feature of COSMO-

RS is the reduction of molecular properties to a probability distribution of screening

charges for the molecule solvated in a perfect conductor, as the sigma profile of the

molecule. Then the sigma profiles are used to estimate the exchange energies between the

substrate and the solvent [6].

The standard procedure of COSMO-RS calculations consisted of two main steps:

quantum chemical calculations and COSMO-RS calculations performed within the ADF

program [7, 8]. Because of the large number of atoms in coenzyme Q10, the density

functional (DFT) optimization of the molecule was very slow by COSMO calculations and

was alternatively carried out by Gaussian03 at the B3LYP/6-311?G level [9]. Then the

resulting files were imported into the ADF program to calculate the total energy of a

coenzyme Q10 molecule in solution and the 3D polarization density distribution on the

molecular surface, i.e., the sigma profile. The COSMO calculations are by the BP density

functional theory with the triple-valence polarized large basis set (TZVP) [10]. The cal-

culations of activity coefficient and solubility were performed with an iterative method,

since the activity coefficient depends on the mole fraction of this compound [11]. The

enthalpy of fusion (112.8 kJ�mol-1) and melting temperature (322.35 K) of coenzyme Q10

Fig. 2 The DSC curve obtained from crystals of Q10. The solid obtained from each solvent showed asimilar DSC pattern. The measurements were made under fixed conditions of constant heating rate of5 K�min-1 and under nitrogen atmosphere (20 mL�min-1)

J Solution Chem

123

was taken as input data for the calculations of solubility. The fusion data were measured by

a NETZSCH DSC 204 differential scanning calorimeter (Fig. 2) and are consistent with

those in the literature [12].

3 Results and Discussion

3.1 Experimental Solubility

The solubility of coenzyme Q10 in ethyl acetate, n-hexane, 1-butanol, 1-propanol, 2-

propanol and ethanol at different temperatures are presented in Table 1 as xexpt. The

experimental data were correlated with the Apelblat equation[13]:

lnðxÞ ¼ Aþ B

T=Kþ C lnðT=KÞ ð2Þ

where x is the mole fraction of coenzyme Q10, and A, B and C are the parameters of the

Apelblat equation. The values of parameters are given in Table 2. The values of the ‘‘A’’

and ‘‘B’’ coefficients for coenzyme Q10 in 2-propanol are very different from those in the

other alcohols. Identifying the mechanism causing this difference requires further research.

Relative root-mean-square deviations (RMSD) were calculated by Eq. 3 [14] and are

also listed in Table 2:

RMSD ¼XN

i¼1

ðx� xcalcÞ2

N � 1

" #1=2

ð3Þ

Fig. 3 Powder X-ray diffraction patterns for the original Q10 (underside) and Q10 dried from clear1-propanol solution (upside). The patterns of Q10 dried from the other solutions in gravimetric experimentsare not shown since they are similar

J Solution Chem

123

where N is the number of data points and x is the saturated solution mole fraction of Q10.

The coenzyme Q10 solubility curves versus temperature in the different solvents are

shown in Fig. 4. This plot indicates that the solubilities in the six pure solvents (ethyl

acetate, n-hexane, 1-butanol, 1-propanol, 2-propanol and ethanol) increased with

increasing temperature as shown in Table 1. At the same temperature, the solubility trend

is ethyl acetate [ n-hexane [ 1-butanol [ 1-propanol [ 2-propanol [ ethanol. The cal-

culated solubility data from the modified Apelblat model is in good agreement with the

experimental data.

3.2 Deviation Between the Experimental and Calculated Solubility

The mole fractions of coenzyme Q10 in solvents were refined using the iterative solubility

calculations of COSMO-RS [6]:

log10ðxSOLðiþ1Þ2 Þ ¼ ½l

ðPÞ2 � lð1Þ2 ðx

SOLðiÞ2 Þ þminð0;DfusGÞ�RT ln 10

ð4Þ

where xSOL2 is the mole fraction of the solute (compound 2) dissolved in the solvent 1 at

saturation, lðPÞ2 is the chemical potential of pure compound 2, lð1Þ2 is the chemical potential

of compound 2 at infinite dilution in the solvent 1, and the ‘‘i’’ means the iterative cycle in

the calculations. Then the solubility of compound 2 in each selected solvent x2 can be

calculated. The calculated solubility in the studied systems are also listed in Table 1 as xpre

and compared with experimental results in Fig. 4. The solubility of coenzyme Q10 was

underestimated by COSMO-RS calculations in ethyl acetate and hexane, while conversely

in 1-butanol, 1-propanol, 2-propanol and ethanol it was overestimated. The difference may

result from an inaccurate calculation of the solute–solvent interactions because the coen-

zyme Q10 molecule has 153 atoms and a very long alkenyl chain, making it difficult to

obtain an accurate sigma profile. Besides, the ability of an alcohol to form a hydrogen bond

with coenzyme Q10 also influences the calculated thermodynamic properties. The

hydrogen bond interaction disturbs the calculation of local electronic polarizability of a

compound molecule while the chemical potential of a compound is a function of its local

polarizability [5].

Although there is a large difference between the experimental data and that predicted by

the COSMO-RS method, the COSMO-RS results displayed a similar temperature depen-

dence of solubility as the measured data. Both of them showed that the solubility of

coenzyme Q10 monotonically increases with temperature in the range considered in this

work.

Table 2 Parameters of the Apelblat equation for coenzyme Q10 in different solvents

Solvents A B C 105 9 RMSD

Ethyl acetate -267.34 -347.45 46.65 2.63

n-Hexane -238.07 -798.80 41.79 3.70

1-Butanol -401.45 266.90 69.03 0.67

1-Propanol -194.93 158.21 32.92 2.61

2-Propanol -318.98 -112.63 54.64 1.31

Ethanol -340.74 122.72 57.86 1.46

J Solution Chem

123

4 Conclusions

The solubility of coenzyme Q10 in different solvents at various temperatures was mea-

sured by a gravimetric method and compared with the predicted data using COSMO-RS.

The solubility increased with temperature and was correlated with the Apelblat equation.

The predicted results showed a similar temperature dependence of solubility compared

with the experimental data. The experimental solubility and correlation equation in this

work can be used as essential data and models in the crystallization process of coenzyme

Q10.

Fig. 4 Experimental and predicted mole fraction of coenzyme Q10 in different solvents at varioustemperatures: in order of ethyl acetate, n-hexane, 1-butanol, 1-propanol, 2-propanol, ethanol

J Solution Chem

123

Acknowledgments We gratefully acknowledge financial support from the National Natural ScienceFoundation of China (No.20836005, 21003077 and 21176184) and the Open Project of Key LaboratoryAdvanced Energy Materials Chemistry (Nankai University) (KLAEMC-OP201201). The ADF program(trial version) was provided by Beijing Hongcam Company.

References

1. Lenaz, G., Fato, R., Formiggini, G., Genova, M.L.: The role of coenzyme Q in mitochondrial electrontransport. Mitochondrion 7(Supplement), S8–S33 (2007)

2. Crane, F.L.: Biochemical functions of coenzyme Q10. J. Am. Coll. Nutr. 20, 591–598 (2001)3. Zhang, J., Hu, J.W., Wang, J.L., Chen, L.Z.: Solubility of 1,3,3-trinitroazetidine in ethanol?water

systems from (293.15 to 323.15 K). J. Solution Chem. 40, 703–708 (2011)4. Jing, D.D., Wang, J.K., Wang, Y.L.: Solubility of penicillin sulfoxide in different solvents. J. Chem.

Eng. Data 55, 508–509 (2010)5. Klamt, A.: COSMO-RS, From Quantum Chemistry to Fluid Phase Thermodynamics and Drug Design.

Elsevier B.V, Amsterdam (2005)6. Schroder, B., Santos, L.M.N.B.F., Marrucho, I.M., Coutinho, J.A.P.: Prediction of aqueous solubilities

of solid carboxylic acids with COSMO-RS. Fluid Phase Equil. 289, 140–147 (2010)7. ADF2010, Trial version, Scientific Computing & Modelling NV, Theoretical Chemistry. Vrije Uni-

versiteit, Amsterdam, Netherlands (2011)8. te Velde, G., Bickelhaupt, F.M., van Gisbergen, S.J.A., Guerra, C.F., Baerends, E.J., Snijders, J.G.,

Ziegler, T.: Chemistry with ADF. J. Comput. Chem. 22, 931–967 (2001)9. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Montgomery,

J.A. Jr., Vreven, T., Kudin, K.N., Burant, J.C., Millam, J.M., Iyengar, S.S., Tomasi, J., Barone, V.,Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G.A., Nakatsuji, H., Hada, M., Ehara, M.,Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene,M., Li, X., Knox, J.E., Hratchian, H.P., Cross, J.B., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann,R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Ayala, P.Y., Morokuma, K.,Voth, G.A., Salvador, P., Dannenberg, J.J., Zakrzewski, V.G., Dapprich, S., Daniels, A.D., Strain, M.C.,Farkas, O., Malick, D.K., Rabuck, A.D., Raghavachari, K., Foresman, J.B., Ortiz, J.V., Cui, Q., Baboul,A.G., Clifford, S., Cioslowski, J., Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I.,Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A., Peng, C.Y., Nanayakkara, A., Challacombe, M.,Gill, P.M.W., Johnson, B., Chen, W., Wong, M.W., Gonzalez, C., Pople, J.A.: Gaussian03, Re VisionC.01. Gaussian, Inc., Pittsburgh (2004)

10. Schafer, A., Huber, C., Ahlrichs, R.: Fully optimized contracted Gaussian basis sets of triple zetavalence quality for atoms Li to Kr. J. Chem. Phys. 100, 5829–5835 (1994)

11. Pye, C.C., Ziegler, T., van Lenthe, E., Louwen, J.N.: An implementation of the conductor-likescreeningmodel of solvation within the Amsterdam density functional package. Part II. COSMO for realsolvents. Can. J. Chem. 87, 790–797 (2009)

12. Zhou, H.F., Yue, Y., Liu, G.L., Li, Y., Zhang, J., Yan, Z.M., Duan, M.X.: Characterisation and skindistribution of lecithin-based coenzyme Q10-loaded lipid nanocapsules. Nanoscale Res. Lett. 5,1561–1569 (2010)

13. Cui, P.L., Yin, Q.X., Gong, J.B.: Solubility of candesartan cilexetil in different solvents at varioustemperatures. J. Chem. Eng. Data 56, 658–660 (2011)

14. Zhang, X.W., Yin, Q.X., Gong, J.B., Liu, Z.K.: Solubility of 5-amino-N,N’-bis(2,3-dihydroxypropyl)-2,4,6-triiodobenzene-1,3-dicarboxamide in ethanol plus water mixtures. J. Chem. Eng. Data 55,2355–2357 (2010)

J Solution Chem

123