Jingzhou Guo ,Yuanzuo Zou ,Bo Shi ,Yuan Pu,Jiexin Wang ,Dan Wang,*,Jianfeng Chen
1 State Key Laboratory of Organic Inorganic Composites,Beijing University of Chemical Technology,Beijing 100029,China
2 Research Center of the Ministry of Education for High Gravity Engineering and Technology,Beijing University of Chemical Technology,Beijing 100029,China
Keywords:Nanonization Solubility enhancement Optoelectronic molecules Solution processing
ABSTRACT Solubility enhancement has been a priority to overcome poor solubility with optoelectronic molecules for solution-processable devices.This study aims to obtain experimental data on the effect of particle sizes on the solubility properties of several typical optoelectronic molecules in organic solvents,including the solubility results of 1,3-bis(9-carbazolyl)benzene (mCP),1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)ben zene(TPBi)and 2-(4-tert-butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole(PBD)in ethanol and acetonitrile,respectively.Nanoparticles of mCP,TPBi and PBD with sizes from dozens to several hundred nanometers were prepared by solvent antisolvent precipitation method and their solubility were determined by using isothermal saturation method.The saturation solubility of nanoparticles of three kinds of optoelectronic molecules exhibited increase of 12.9%-25.7% in comparison to the same raw materials in the form of microparticles.The experimental evidence indicates that nanonization technology is a feasible way to make optoelectronic molecules dissolve in liquids with enhanced solubility.
Over the past decades,organic light-emitting diodes (OLEDs)are garnering an increasing amount of research interest because of their high contrast ratio,color purity,and low power consumption[1-4].Vacuum vapor deposition has been the main process for depositing OLED materials.However,the high cost of the process,including the expensive vacuum equipment and the waste of material during vapor deposition,are hindering the widespread application of the technology particularly for large-size OLED displays [5,6].Alternatively,solution-processable roll-to-roll electronic devices have been extensively developed,which significantly enhance the material utilization efficiency and reduce the manufacturing costs [7-9].To achieve the goal to print large size OLED displays,the functional materials for light-emitting layer and electron transport layer have to be converted into printable inks for solution processing,while most of the currently used OLED materials are poorly soluble in organic solvents [10-14].To date,the efforts to enhance dissolution of optoelectronic molecules mainly focus on the approaches to graft soluble groups to increase the solubility[15,16].Although designing new molecules and new materials has the advantages of high-performance and multiaccessibility,its shortages of high development cost and long development period are inevitable.Nanonization is the process of transforming poorly soluble molecules into nanoparticles with high surface area,making them easily soluble in liquids [17-19].This technique has been widely recognized on drug discovery and development[20-23].According to the fundamental equation of thermodynamics,the thermodynamic binding affinity is dependent on the binding enthalpy and entropy[24].Based on the analysis of thermodynamics mechanism,the nanonization of insoluble OLED materials offers a feasible way to make existing optoelectronic molecules dissolve in liquids with enhanced solubility.On the other hand,the improved efficiency of solubility can reduce the processing time for devices and guarantee their quality[25,26].However,systematic experimental study on the effect of particle sizes on the solubility properties of optoelectronic molecules in organic solvents is absent and the quantitative experimental data is even less,which hinder the development of the nanonization technology that can produce higher labor productivity and better economic results.
Aiming to fill this gap,experimental data on the effect of particle sizes of three kinds of commonly used optoelectronic molecules including 1,3-bis(9-carbazolyl)benzene(mCP),1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)benzene (TPBi) and 2-(4-tert-butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole (PBD) are investigated in this work.The mCP is a kind of typical materials for OLED lightemitting layer with a simple molecular structure and high triplet state energy,while TPBi and PBD are widely used electron transport materials with high electron mobility and wide energy gap.The nanoparticles of mCP (~600 nm),TPBi (~170 nm) and PBD(~50 nm)are prepared by using the solvent antisolvent(SAS)precipitation method.The nanoparticles are characterized by Fourier transform infrared spectroscopy (FTIR),ultraviolet and visible spectrophotometry (UV-Vis),as well as scanning electron microscopy (SEM) to investigate the morphological changes of the precipitation products of the three materials under different experimental conditions.The dissolution equilibrium curves of the three materials in various solvents are plotted and the dissolving characteristics of the nanoscale materials and raw materials are investigated.
The mCP(mass fraction purity ≥97.0%),TPBi(mass fraction purity ≥97.0%)and PBD(mass fraction purity ≥97.0%)were purchased from Sigma-Aldrich,Germany.The organic solvents including ethanol,acetonitrile,N-methylpyrrolidone (NMP) and tetrahydrofuran (THF) were all analytical pure and were used as received.The details of the chemicals were summarized in Table 1.The UV-Vis absorption spectra were measured usinga Shimadzu UV-2600 spectrophotometer.The FTIR spectra were recorded using a Bruker Vector-70v FTIR spectrometer and SEM images were taken by a JEOL JSM-7800F scanning electron microscope.
The solubility of mCP,TPBi and PBD in ethanol and acetonitrile were determined by isothermal satiation method [27] at 20 °C under atmospheric pressure,respectively.Briefly,taking the determination for mCP solubility in ethanol as the example,a series of mCP solutions with known concentrations were prepare and the UV-Vis absorbance spectra of the solutions were measured.The calibration equation of mCP in ethanol was then calculated according to the absorbance data and known concentration points.Then,the saturated solution of mCP in ethanol was prepared by adding excess mCP into 10 ml screw cap vial containing ethanol,following by sonicating for 15 min in water bath with 20 °C.The mixture was then stirred for 1 h with an electromagnetic stirrer.It should be ensured that there was still a surplus solid-phase visible during the operation,indicating that the solution was saturated.The solution was then allowed to be kept at the same temperature for 2 h,and the excess solid phase was filtered using a Teflon membrane to obtain a saturated solution.The absorbance of saturated solution was measured and the saturated solubility was obtained through calculation by the Lambert-Beer law.
For each type of materials,three samples of nanoparticles with different morphologies and sizes were preparedviasolvent antisolvent (SAS) precipitation method in a rotating packed bed(RPB) reactor,which consisted of a packed rotator,a seal ring,an inlet and an outlet [28,29].Typically,taking the preparation of mCP nanoparticles as an example,the mCP molecules were firstly dissolved in NMP to form the solvent phase.The solvent of mCP and antisolvent (water) were pumped with a fixed volume ratio of 1:10 into the RPB reactor working at 2540 r.min-1.Due to the sudden increase in the supersaturation of the mixed solution,the mCP in the solution rapidly formed precipitate.The resulting mixture was collected at the outlet,and was filteredviaa 0.2 μm filter membrane to get precipitation of mCP.The precipitated products were then vacuum dried for further testing and characterization.By choosing THF as the solvent,the nanoparticles of TPBi and PBD were prepared following the similar method.
The dissolution equilibrium curves and saturation solubility of three kinds of nanoparticles in ethanol and acetonitrile at 20 °C were measured and analyzed.Briefly,the excess powders of nanoparticles were added to the vial containing the organic solvent.A small amount of the solution was taken at different time points and quickly filteredviaa 0.2 μm filter membrane.The concentration of the solutes in the filtrate was determined as the solubility of the optoelectronic molecules in the form of nanoparticles in the organic solvent at that time point.
Fig.1(a) shows the chemical structure of mCP,TPBi and PBD.The mCP has been a hot research object for vacuum deposition of phosphorescent OLEDs,while both TPBi and PBD are usually used as electron-transporting material in devices of OLEDs.The knowledge on the solubility enhancement of these optoelectronic molecules in solvents is therefore important for optimizing operation condition and enlarging equipment to study solution-processable electronic devices.To obtain the experimental data of the solubility,the UV-Vis spectra of these solutes in ethanol and acetonitrile were firstly measured.As the results shown in Fig.1(b)and(c),the absorption peaks of mCP were observed at 292 nm and 291 nm in ethanol and acetonitrile,respectively.The absorption peaks of TPBi in ethanol and acetonitrile were located at 300 nm and 302 nm,respectively.The absorption peak of PBD in ethanol and acetonitrile were around 305 nm and 303 nm,respectively.The slightly different peaks wavelength was attributed to the dispersion interactions of the materials in organic solvents.

Fig.1.(a) Chemical structure of 1,3-bis(9-carbazolyl)benzene (mCP),1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)benzene (TPBi) and 2-(4-tert-butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole(PBD).UV absorption spectra of mCP,TPBi and PBD in(b)ethanol and(c)acetonitrile,respectively.The concentration of the solute was 0.005 mg.mL-1 for all samples.
The absorbance correspondingAmaxto the peak wavelength is used to calculate the concentration following the Lambert-Beer law [30] in Eq.(1).
whereCis the molarity concentration of the solution and ε is molar absorption coefficient of materials.The calibration equation for each solvent is derived from the absorbance data of a series of known concentration points using Eq.(2),based on the proportional connection between absorbance and concentration.
in which Abs means the absorbance,Con is the known concentration of solution,andAandBare the slope and intercept of the calibration equation.The mole-fraction solubility of mCP,TPBi and PBD is calculated by using Eq.(3).
wherexerepresents the mole fraction solubility of the solute,m1andm2are the masses (g) of the solute F6-Br and solvent,respectively.M1andM2are the molar mass (g.moL-1) of the solute and solvent.Accordingly,the calibration equations for the solubility measurement of mCP,TPBi and PBD in ethanol and acetonitrile is summarized in Table 2.

Table 2 The calibration equations and molar fraction solubility x of mCP,TPBi and PBD in solvents at p=0.1 MPa
Fig.2 shows the SEM images of the raw materials and nanoparticles of mCP,TPBi and PBD prepared by a well-developed solvent antisolvent precipitation method in high-gravity-assisted RPB reactor.The raw materials were irregular lumps with a radius of more than 10 μm,while TPBi and PBD were particles with radius of 100 μm or more (Fig.2(a)-(c)).After processing of SAS precipitation,the mCP nanoparticles exhibited tiny spheres with average particle size of(600± 100)nm (Fig.2(d)).The processed TPBi presented an irregular polygon and the particle size arranged from 150 nm to 200 nm (Fig.2(e)).The PBD nanoparticles also showed an irregular polygon,while the particle size was obviously smaller than 100 nm.According to the average diameter analysis,the TPBi nanoparticles showed average size of(170±20)nm(Fig.2(e))and the average size of PBD nanoparticles was around 50 nm(Fig.2(f)).

Fig.2.SEM images of raw materials of (a) mCP,(b) TPBi,(c) PBD,and corresponding nanoparticles of (d) mCP,(e) TPBi,(f) PBD,respectively.
The FTIR spectra of raw materials and nanoparticles of mCP,TPBi and PBD were present in Fig.3.In Fig.3(a),the absorption peak at 3060 cm-1was caused by the C-H stretching vibration of benzene ring.The two strong peaks at 1600 cm-1and 1496 cm-1were due to the stretching vibration of the C=C.Absorption vibration band at 751 cm-1indicates the existence of a 1,2-disubstituted benzene ring[31].In Fig.3(b),the C—H stretching vibration peak of benzene ring was observed at 3060 cm-1,while the absorption peaks at 1590 cm-1and 1448 cm-1were attributed to the C=N stretching vibration.In Fig.3(c),the absorption peaks near 3030 cm-1and 1610 cm-1indicated the C—H stretching vibration and C=C stretching vibration of benzene ring.The absorption peaks at 2962 cm-1and 1480 cm-1corresponded to the asymmetrical stretching vibration and formation vibration of the saturated alkane terminal CH3,and 850 cm-1and 733 cm-1in the fingerprint region corresponded to the C—H outof-plane formation vibrations [32].It was noted the fingerprint patterns in the FTIR spectra of nanoparticles for each kind of molecules were consistent with those of the raw materials.These results indicated that the nanonization processing did not affect the chemical structure of mCP,TPBi and PBD molecules.

Fig.3.FTIR spectra of raw materials (black line) and nanoparticles (red line) of (a) mCP,(b) TPBi and (c) PBD,respectively.
The solubilization process of poorly soluble optoelectronic molecules in liquid usually comprises several steps [33,34] which is shown schematically in Fig.4(a)-(c).Firstly,the intermolecular attraction forces must be overcome for a molecule to be removed from the solute phase to the solvent.The molecules of the solvent,that is either acetonitrile or ethanol in this study,are reorganized as a function of the environment dipole(Fig.4(a))with a tendency to migrate to the solute phase,which detaches(Fig.4(b))and finds a cavity in the solvent large enough to receive the solute molecule(Fig.4(c)).The saturation solubility is a constant that depends on the nature of the material,the temperature at which it dissolves,and the crystal structure and particle size.In particular,the saturation solubility increases when the particle size decreases below 1 μm.This effect can be explained by the Ostwald-Freundlich equation (Eq.(4)),which describes the relationship between the saturation solubility of the material (Cs) and the particle size (r)[35].

Fig.4.Graphic representation of the solubilization process:(a)migration of solvent molecules,(b)detachment of the solute phase molecule and(c)formation of the cavity in the solvent phase to accommodate the solute.(d) Saturation solubility as a function of r and γ.
whereCsis the saturation solubility,C∞is the solubility of large particles (r→∞),γ is the interfacial tension of the substance,Mis the molar mass of the particle material,ρ is the density of the solid,Ris the gas constant,Tis the absolute temperature,andris the radius.Fig.4(d) shows that the saturated solubility is determined by particle sizerand interfacial tension γ.The values of γ depends on the nature of the substance and the choice of solvent,The γ values without any absorb foreign molecules are typically between 5-50 according to the previous literature [36].
As mentioned above,the raw materials of mCP,TPBi and PBD exhibited irregular lumps and particle sizes of more than 10 μm or over 100 μm.It was hard to quantify and completely described the increasing in solubility for nanoparticles by Eq.(4).However,with qualitative analysis on the basis of solution equilibrium,it tells that the formation of nanodispersions of nanoparticles in solvents is also an important reason for the solubility enhancement.As the diagram shown in Fig.5,upon addition of nanoparticles to the solvent,the uniform size distribution of particles in the solvents results in equilibrium of dissolution and diffusion in the nanometer to micrometer scale.The growth of particles is restricted and the average particle size remains unchanged in certain time scale,forming transparent nanodispersion of the solutes along with molecular dissolution.Thus the apparent solubility of the nanoparticles of the solute is enhanced compared with that of raw material.

Fig.5.Graphic representation of the solubilization process of micro-scale particle and aggregate of nanoparticles,respectively.
Fig.6 shows the dissolution equilibrium curves of mCP,TPBi and PBD raw materials and nanoparticles in ethanol and acetonitrile,respectively.The saturation solubility of nanoparticles of mCP in ethanol and acetonitrile was enhanced by 22.9% and 12.9% compared to the raw materials;the saturation solubility of nanoparticles of TPBi in ethanol and acetonitrile was enhanced by 19.6%and 25.7%compared to the raw materials;the saturation solubility of nanoparticles of PBD in ethanol and acetonitrile was enhanced by 13.3% and 17.4% compared to the raw materials.
Nanonization enhanced solubility for poorly soluble mCP,TPBi and PBD compared to their raw materials can be explained by the Noyes-Whitney equation (Eq.(5)):
where dc/dtis dissolution rate,Dis diffusion coefficient,Ais surface area,his the diffusion layer thickness,Csis the saturation concentration,Ctis the concentration at timet.Under the condition of constant temperature,Dandhremain constant,whileAvaries due to the influence of the dissolution process [37].ThereforeAshould be considered as a function of timet.Since the raw materials of mCP,TPBi and PBD exhibited irregular lumps and particle sizes,it was hard to quantitative analysis the experimental results of raw materials and nanoparticles according to Eq.(5).However,it could be clearly understood that the effect of nanonization enhanced solubility of OLED raw materials and intermediate materials was beneficial for dissolution technical-economical indexes improvement.For example,Fig.7 presents the disperse and/or dissolve processing of a kind of typical insoluble molecule(2.5 mg) in NMP (1 ml) under ultrasonic treatment from 0 to 60 s.It was obvious that the nanoparticles were easily dissolved,forming a stable and clear solution (Fig.7(a)).However,the raw materials were hardly dissolved,forming turbid liquid with NMP (Fig.7(b)).The mixtures in two bottles were irradiated by a red laser to investigate the Tyndall effects.No obvious scattering of particles were observed from the solution of nanoparticles (the left bottle in Fig.7(c)),indicating that the molecules were dissolved and existed in the molecule state or ultrasmall aggregate.The strong scattering of red light in the mixtures of raw materials dissolved in NMP (the right bottle in Fig.7(c) demonstrated that the insoluble molecules were not fully dissolved and existed in the state of large particles.

Fig.7.The disperse and/or dissolve processing of (a) nanoparticles and (b) raw materials of same insoluble molecule with same amount (2.5 mg) in NMP (1 ml) under ultrasonic treatment from 0 to 60 s.(c)The mixtures of nanoparticles dissolved in NMP(left bottle)and raw materials dissolved in NMP(right bottle)irradiated by a red laser.
The aim of this study is to investigate the effect of particle size on the solubility of optoelectronic molecules in organic solvents.Dissolution equilibrium curves of nanoparticles of mCP(~600 nm),TPBi(~170 nm)and PBD(~50 nm)and corresponding raw materials in the size of several micrometers are determined.The experimental results showed that the nanoparticles of the three kinds of molecules showed significant increasing in saturation solubility of 12.9%-25.7% compared with those of the raw materials.In practical industrial production process,dissolution of some kinds of insoluble molecules in in some specific solution for reaction has been a time consuming and labor intensive process.The experimental verification of nanonization enhanced solubility for poorly soluble optoelectronic molecules provides a basis and thinking method to solve the difficulty of dissolving organic optoelectronic materials in organic solvents.By using pre-formed nanoparticles of the insoluble molecules,the energy consumption and operating cost for dissolution could be reduced from the perspective of system operation.As the proverb says,‘‘Sharpening the axe will not delay the job of cutting wood”.More quantitative analysis,including the particle sizes and morphologies on the solubilities and dissolution rates,are helpful to understand the theoretically predicted trends and accurate models,which should be good topics for future studies.
Data Availability
Data will be made available on request.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We are grateful for financial support from National Natural Science Foundation of China (22288102) and the Fundamental Research Funds for the Central Universities of China(buctrc202016).
Chinese Journal of Chemical Engineering2023年8期