- Email: [email protected]
Mechanistic study on rapid fabrication of fibrous films via centrifugal melt spinning
International Journal of Pharmaceutics 560 (2019) 155–165
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Mechanistic study on rapid fabrication of fibrous films via centrifugal melt spinning ⁎
Yan Yanga,b, Nan Zhenga, Yanjun Zhoua, Weiguang Shana, , Jie Shenb, a b
T
⁎
College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, China College of Pharmacy, University of Rhode Island, Kingston, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Centrifugal melt spinning (CMS) Fibrous films Melt rheology Fiber formation mechanism Computational fluid dynamics Flow field analysis
Fibrous films have attracted considerable attention in the field of drug delivery and wound dressings owing to their porous structure and highly aligned fiber orientation. However, current fabrication methods such as electrospinning have certain limitations, including high voltage requirement and conductivity dependency. This has greatly hindered the product development and applications of fibrous films. The objective of the present study was to develop a high throughput and solventless fiber fabrication method via centrifugal melt spinning (CMS) technology. A mechanistic study on the rapid fabrication of drug-loaded fibrous films was conducted using different model drugs and polymers. It was observed that the formability, morphology, and yield of fibrous films were affected by melt rheological properties of film components, operation temperature, and plasticizers. Maintaining suitable fluidity of molten materials during the CMS process is critical for the fiber formation. The produced fibrous films had high drug loading, highly aligned orientation and modulatable drug dissolution characteristics. Finally, computational fluid dynamics (CFD) was used to simulate the melt flow fields during the CMS process. Pressure, turbulence, velocity, and partial pathlines were simulated to elucidate the influence of various operation parameters (i.e. rotating speed, inlet rate and collecting radius) and material properties (i.e. density and viscosity) on the outlet velocity of products and sample collection position. The present study demonstrated that CMS is a high throughput and cost-efficient fabrication method for drug-loaded fibrous films. CFD simulation can be used to assist in understanding fiber formation as well as optimization of CMS process parameters.
1. Introduction Fibrous films have gained widespread interest in the fields of pharmaceutics as drug delivery systems and bio-functional scaffolds, owing to their high surface area and drug loading capability, and highly porous structure and aligned fiber orientation. For example, fibrous film-based oral fast-dissolving films have been successfully produced (Illangakoon et al., 2014; Li et al., 2013). In addition, fibrous films have been used as functional wound dressings since they can absorb excess exudates, ensure sufficient gas and nutrient exchange and promote cell proliferation and migration (Yang et al., 2017a,b). Electrospinning is commonly used to produce fibrous films, in which a film component solution or molten material is (are) stretched into fibers via high voltage electrostatic force. Fibrous films based on pharmaceutical polymers such as polyvinylpyrrolidone (PVP) (Illangakoon et al., 2014), polyvinyl alcohol (PVA) (Li et al., 2013),
PVA/chitosan (Yan et al., 2014), poly(lactic-co-glycolic acid) (PLGA) (Wang and Suo, 2012), ethyl cellulose (EC) (Li et al., 2014) and cellulose acetate (Yan et al., 2013), have been produced using electrospinning. However, electrospinning technology has certain limitations, including high voltage (> 10 kV) requirement and conductivity dependency (Zander, 2015). An alternative approach to produce ultrafine fibers is centrifugal melt spinning (CMS), which employs centrifugal force to stretch the melt jet from the spinneret (Zander, 2015). CMS offers a solventless and high throughput fiber formation approach. It also enables fiber formation of polymers and/or drugs with poor or nonconductivity and hence broader applications. Most recently, CMS has been used to prepare sucrose-based microfibers containing poorly water-soluble drugs (i.e. olanzapine and piroxicam) (Marano et al., 2016). The drugs were found to be present in the amorphous state, and drug in vitro dissolution performance was significantly enhanced. Further stability study revealed that even aged fibers (freely recrystallized)
⁎ Corresponding authors at: College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, China, Tel.: +86-0571-88871075 (W. Shan) and College of Pharmacy, University of Rhode Island, Kingston, RI, United States Tel.: +1-401-874-5594 (J. Shen). E-mail addresses: [email protected] (W. Shan), [email protected] (J. Shen).
https://doi.org/10.1016/j.ijpharm.2019.02.005 Received 19 December 2018; Received in revised form 2 February 2019; Accepted 4 February 2019 Available online 12 February 2019 0378-5173/ © 2019 Elsevier B.V. All rights reserved.
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Nomenclature PVP PVA PLGA EC CMS CFD IBU TNZ MT NF IND PEG EPO RL RS SOL VEA
TEC GTA Tm Td TGA DSC η* η′ XRD MRF ρ Vin ω R △T1 △T2 Ts θ
polyvinylpyrrolidone polyvinyl alcohol polylactic-co-glycolic acid ethyl cellulose centrifugal melt spinning computational fluid dynamics ibuprofen tinidazole metoprolol tartrate nifedipine indomethacin polyethylene glycol Eudragit® EPO Eudragit® RL PO Eudragit® RS PO Soluplus® vitamin E acetate
triethyl citrate glycerol triacetate melting temperature thermal decomposition temperature thermal gravimetric analysis Differential scanning calorimetry complex viscosity dynamic viscosity X-ray diffraction multiple reference frame density inlet velocity rotational speed entrance radius melt operating temperature range fiber-forming temperature range soft temperature rotation angle
prepared using an in-house device, and various drugs and pharmaceutical polymers were tested for their fiber formability. The effects of film components, operation temperature and plasticizers on the formability, morphology, and yield of drug-loaded fibrous films were investigated. Melt rheology was employed to explore the thermoforming mechanism of film components (i.e. drugs with/without polymers). Lastly, CFD was implemented to simulate and assess the melt flow field based on twodimensional plane projection models.
retained dissolution enhancement capability (Marano et al., 2017). Despite of the fact that CMS has been shown potential in manufacturing drug-loaded fibers, the literature on the one-step fabrication of fibrous films via CMS has remained sparse. Commercial centrifugal spinning devices such as Cyclone Fiber Engine (FE 1.1M/S), Cyclone L1000 and cotton-candy machines can only be used to prepare fibers, not fibrous films and additional step(s) is(are) often needed to fabricate films from fibers, which is difficult to control and not cost effective. External collectors such as rotary drums (Zhou et al., 2018) and round bottom collectors (Loordhuswamy et al., 2014) have been used during the CMS process to collect fibrous scaffold. Till now, whether CMS can be used for one-step formation of drug-loaded fibrous films; and whether fibrous films can be used to achieve sustained drug release have not been studied or reported. Furthermore, fiber formation mechanism via the CMS technology has yet to be understood. Computational fluid dynamics (CFD) is a branch of fluid mechanics that uses numerical analysis and data structures to simulate the interactions of liquids with surfaces defined by boundary conditions. CFD has been used to simulate and analyze thermoforming techniques. For example, the airflow field of melt-blowing process was simulated using CFD (Hassan et al., 2016), and the results were used to improve die configurations to prepare melt blown fibers with reduced fiber diameter and pore size. In addition, CFD was used to simulate the jet trajectory of a spinning wheel atomizer (Mencingeret al., 2015) to investigate the influence of process variables on the ligament growth. Therefore, it was envisioned that CFD simulation might be a feasible way to visualize and analyze the fiber formation process during CMS and assist in the process and formulation optimization. The present study aims to conduct a mechanistic study on the rapid fabrication of drug-loaded fibrous films via CMS. Fibrous films were
2. Materials and methods 2.1. Materials Ibuprofen (IBU) (No. 20151224) was obtained from Zhenyuan Chemical Co., Ltd. (Hubei, China). Tinidazole (TNZ) (No. 164862), metoprolol tartrate (MT) (No. 131069) and nifedipine (NF) (No. 164862) were purchased from YuanCheng Pharmaceutical Co., Ltd. (Hubei, China). Indomethacin (IND) (No. 20160103) was supplied by XiYinHe Chemical Co., Ltd. (Hubei, China). Polyethylene glycol (PEG) (No. 60111, Mw = 6000) was purchased from XiLong Chemical Co., Ltd. (Guangxi, China). Eudragit® EPO (EPO) (No. G150131502), Eudragit® RL PO (RL) (No. G080936208) and Eudragit® RS PO (RS) (No. G111238239) were supplied by Evonik Rohm Co., Ltd. (Germany). Soluplus® (SOL) (No. 84414368EO) was provided by BASF (Germany). EC (Std. 20, No. PD403586) were obtained from Colorcon Coating Technology Co., Ltd. (Shanghai, China). Vitamin E acetate (VEA) (No. VA160404) was obtained from Healthful Biotechnology Co., Ltd. (Xian, China), triethyl citrate (TEC) (No. 150605) was obtained from FengyuanTushan Pharmaceutical Co., Ltd. (Anhui, China) and glycerol triacetate (GTA) (No. 20160623) was supplied by SuZhong Technology
Fig. 1. Illustration of the in-house centrifugal melt spinning (CMS) device. (a) Perspective view; and (b) top view. 156
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Greifensee, Switzerland) (Yang et al., 2018). IBU, TNZ, and MT were studied from 25 °C to 150 °C at a rate of 10 °C/min and IND and NF were tested from 25 °C to 200 °C at a rate of 10 °C/min. Thermal behavior of the model drugs was studied.
Table 1 Validated UV calibration curves for IND, NF and TNZ. Drugs
Wavelength
Equations
R
IND NF TNZ
320 nm 333 nm 317 nm
A = 0.0193C A = 0.0149C − 0.0048 A = 0.0362C − 0.0052
1.0000 0.9999 0.9999
2.4. Melt rheological analysis Melt rheology of the model drugs and polymers (i.e. PEG, SOL, and RL) were determined using an Anton Paar Physica Rheometer (MCR 302, Anton Paar, Austria) with a 25.0 mm parallel plate in an oscillation mode and a gap distance of 0.3 mm (Maru et al., 2011). Following the preliminary study, the frequency was set at an angular frequency of 10 rad/s with a strain amplitude of 0.5%. Complex viscosity (η*) and dynamic viscosity (η′) were recorded every 5 s. The equilibrium temperatures were 100 °C for IBU, 150 °C for MT, 180 °C for IND, 190 °C for NF, 100 °C for PEG, and 200 °C for both SOL and RL. The test was conducted at a cooling rate of 0.8 °C/min. The effect of heating and cooling process on the melt rheology of the film components was also investigated.
Chemical Co., Ltd. (Guangdong, China). 2.2. Preparation of fibrous films Fibrous films were prepared using an in-house CMS device (Fig. 1) composed of a melting chamber with a radius of 60 mm. There were 24 outlet sectoral channels (thickness 5.0 mm, 6.0 rad) and each with 3 × 3 spinneret orifices (thickness 0.5 mm, 0.5 rad) on the side wall of the melting chamber. The operation temperature of the melting chamber was precisely controlled via an infrared temperature sensor and a heating device, and adjusted for different film components. The film components (e.g. drugs or drug/polymer mixture) were mixed in a mortar for 10 min and fed into the melting chamber. The melting chamber was then rotated at a speed of 3000 rpm and products were collected in the concentric barrels (as shown in Fig. 1). Morphology and size of the collected products were determined using a stereomicroscope (ZY-HD1400, ZongyanWeiye, China) with S-EYE v1.1 measuring software.
XRD analysis of the products was performed in an X-ray diffractometer (D/max 2550/PC, Rigaku, Japan) with Cu Kα1 radiation (Shimada et al., 2018). Samples were scanned at 40 kV and 40 mA, in the 2θ range of 5–50° at 10°/min.
2.3. Thermal analysis
2.6. In vitro dissolution testing
Thermal analysis was conducted to determine melting temperature (Tm) and decomposition temperature (Td) of the model drugs (i.e. IBU, TNZ, MT, IND, and NF). To determine the Td, thermal gravimetric analysis (TGA) was conducted using a TG analyzer (Q5000, TA instrument, New Castle, USA) (Khan et al., 2016). Samples were heated from 25 °C to 400 °C at a rate of 10 °C/min under a nitrogen atmosphere, and the Td of the model drugs was determined. In addition, approximately 5 mg of the model drugs were weighed and analyzed using a differential scanning calorimeter (DSC1, Mettler-Toledo,
In vitro dissolution of the fabricated films was conducted using a paddle method (USP Apparatus II, 100 rpm, 37 ± 0.5 °C, 500 mL) (Yang et al., 2018). The release medium was PBS (pH 7.2) for IND, 0.5% SLS solution for NF, and deionized water for TNZ, respectively. At predetermined time intervals, 5 mL of dissolution media were withdrawn and replaced with fresh media. The release samples were filtered through 0.45-μm Millipore filters, and analyzed using a spectrophotometer (UV-2450, Shimadzu, Japan). The UV analytical methods were established and validated for the representative model drugs (i.e.
2.5. X-ray diffraction (XRD) analysis
Fig. 2. (a) Thermo gravimetric analysis (TGA) curves; (b) Differential scanning calorimetry (DSC) diagrams; and (c) Melting temperature (Tm) and melt operating temperature range (△T1) of ibuprofen (IBU), tinidazole (TNZ), metoprolol tartrate (MT), indomethacin (IND) and nifedipine (NF).
157
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
IND, NF, and TNZ) and the calibration curves are shown in Table 1. The cumulative drug release profiles of model drugs were determined. 2.7. Melt flow field simulation Melt flow field simulation was conducted using a two-dimensional plane projection model (Wang et al., 2017) that was meshed using Gambit 2.4 software. The flow domain was divided into a moving zone and a static zone, and Tri&Pave elements were generated using a smart mesh method. Steady simulation of the melt flow fields during the CMS process was conducted using Fluent 6.3 software. A multiple reference frame (MRF) model was used to simulate the axial flow and a standard κ-ε model was adopted to describe the turbulence. The PRESTO! algorithm was used to calculate pressure and the SIMPLE algorithm was used to couple pressure and velocity (Jafarzadeh et al., 2011). The density (ρ) of melts was determined using a Melt Indexer (XNR-400A, Jinhe Instrument, China), while the η′ of melts was determined based on the melt rheological analysis as described above (see 2.4). The inlet velocity (Vin) was calculated using the following equation:
vin = R × ω
Fig. 4. Influence of temperature on complex viscosity (η*) of ibuprofen (IBU), metoprolol tartrate (MT), indomethacin (IND) and nifedipine (NF).
of IND and NF didn’t change significantly until the temperature was 20–30 °C below the Tm of these two drugs (i.e. 162 °C for IND and 174 °C for NF). As a result, the molten IND and NF maintained fluidity to some extent and were stretched into fibers. In the case of MT, the η* value gradually increased when the temperature was below 126 °C (Tm of MT) and drastically surged when the temperature was below 108 °C, indicating that the fluidity of the molten MT was not sufficient for fiber formation. Most of the molten MT solidified into powders with a merely small part of short fiber formation. The effect of operating temperature on fiber formation was also investigated using the two model drugs (i.e. IND and NF) with good fiber formability. The fiber-forming temperature range (△T2) within which fibrous films can form via CMS was 162–186 °C for IND and 174–188 °C for NF, respectively. Unexpectedly, powder formation was observed at high operating temperature (beyond the △T2). Since the η* of the molten IND or NF was too low when the operating temperature was beyond the △T2, the melt jet may break into droplets due to centrifugal force and surface tension and hence powders were formed rather than fibers. These results demonstrated that melt rheology study is critical to determine suitable operating temperature during the CMS process.
(1)
where ω is the rotational speed and R is the entrance radius. 3. Results and discussion 3.1. Pure drug fibrous films 3.1.1. Fiber formability of different drugs The Td determined via TGA was found to be IBU < MT < NF < TNZ = IND (Fig. 2a), while DSC results showed that the Tm was IBU < MT < TNZ < IND < NF (Fig. 2b). Accordingly, the melt operating temperature range (△T1) between Tm and Td was NF < MT < IND < IBU < TNZ (Fig. 2c). The fiber formability of different model drugs was evaluated by setting their Tm as respective operating temperature. Following the fiber formation process, only IND and NF fibrous films (diameter: 5–20 μm) were obtained between the inner and outer collecting barrels (Fig. 3). In the case of IBU and TNZ, powders (size: 50–200 μm) were collected on the outer barrel, whereas a mixture of powder and short fibers were obtained for MT. Melt rheology study was conducted to understand the fiber formability of the model compounds studied. As shown in Fig. 4, all drugs (i.e. IBU, MT, IND, and NF) had a low η* (< 1 Pa·s) at high temperature (near or above its Tm) and the fluids behaved differently during cooling. The η* value of IBU drastically increased as soon as the temperature was below 70 °C. This indicated that the η* of the fluid may be too high to be stretched into fibers when the molten material left the spinneret orifices and hence IBU powders were obtained. In contrast, the η* value
3.1.2. Immediate release characteristics As shown in Fig. 5a, the cumulative drug released from the IND or NF fibrous films was over 90% at 10 min, and much higher dissolution rates and extent were obtained compared with the drugs alone. The XRD results in Fig. 5b showed that IND was in amorphous state and the crystallinity of NF reduced significantly in their respective fibrous films. Accordingly, the fast dissolution rates can be explained by not only the highly porous structure and high surface area of the fibrous films, but
Fig. 3. Products collected on the concentric barrels and their morphology observed using stereomicroscope. IBU: ibuprofen; TNZ: tinidazole; MT: metoprolol tartrate; IND: indomethacin; and NF: nifedipine. 158
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 5. (a) In vitro dissolution profiles and (b) XRD results of indomethacin (IND), IND fibrous film, nifedipine (NF) and NF fibrous film.
increased gradually when cooled from 200 °C to 100 °C. Similar to the fiber formation of pure drug discussed above (see 3.1.1), the moderate η* value facilitated the stretching of the molten polymers into fibers. In contrast, the η* value of PEG drastically increased and the melt jet of PEG solidified into powder without stretching. A high operating temperature was also tested for PEG. It was observed that the melt jet of PEG broke into droplets due to its low viscosity. Therefore, it seemed that only polymers possessing suitable fluidity (gradually increased η* value over a wide temperature range) were good candidates for fibrous film formation via CMS. In addition, the influence of a typical heating and cooling process on the melt rheology was studied using RL as the model polymer. As shown in Fig. 8b, viscosity hysteresis was observed during the heating and cooling process. The phenomenon was related to crystal melting and transformation of polymer, indicating that the internal structure of the polymer may have been changed during thermal operation (O’Haire et al., 2016).
also changes in the solid state of the model drugs. 3.2. Drug-loaded polymer fibrous films 3.2.1. Fiber formability of different polymers The fiber formability of commonly used pharmaceutical polymers was investigated in order to tailor drug release from fibrous films. The minimum operating temperature during CMS was set to be Tm for crystalline polymer (i.e. PEG) and soft temperature (Ts) for amorphous polymers (i.e. EPO, SOL, RL, RS, and EC), respectively. As shown in Fig. 6, the Tm or Ts was PEG < EPO < SOL < RL = RS < EC. Base on the reported Td of different polymers (RL = RS < EC < EPO = SOL < PEG) (Altamimi and Neau, 2018; Grymonpré et al., 2017; Gioumouxouzis et al., 2018; Fujimori et al., 2002; DavidovichPinhas et al., 2014), the △T1, which was the temperature range from Tm or Ts to Td, was RL = RS < EC < SOL < EPO < PEG. The fiber formability of different polymers was evaluated by setting their Tm or Ts as the operating temperature (Fig. 7). Among the six polymers investigated, PEG formed powders with a diameter of 50–200 μm during the CMS process. SOL and EPO formed fibers with diameters of approximately 260 μm and 130 μm, respectively. The larger diameter of SOL might be due to its higher melt viscosity. In spite of a narrow △T1 of 10 °C (Fig. 6), RL and RS also formed uniform fibers with a diameter of approximately 190 μm. Considering that the film components only exposed to high temperature at a very short period of time, production of fibrous films via the CMS process with the instantaneous fiber formation is feasible for thermosensitive materials such as RL (Marano et al., 2016). In addition, EC fiber with a diameter of approximately 20 μm was obtained. Small fiber diameter may has resulted from its low melt viscosity at Ts of 170 °C. The slightly yellow appearance of EC fibers suggested that EC might be oxidized in air at high temperature during the CMS process. Melt rheology study was conducted to explain the different behaviors of the molten polymers (i.e. RL, SOL, and PEG) after leaving the spinneret orifice. As shown in Fig. 8a, the η* value of RL and SOL
3.2.2. Influence of plasticizer type and concentration on fiber formation It was reported that addition of plasticizers such as DBS and TEC can lower the melt processing temperature (Gioumouxouzis et al., 2018; Yang et al., 2017a,b) and refine the fibers by lowering the viscosity of polymer melts (Dalton et al., 2007). The effect of plasticizer type and concentration on Ts of EPO, SOL and RL was evaluated. As shown in Fig. 9a, the addition plasticizer reduced the Ts of polymers, and GTA (20%, w/v) was the best plasticizer among the plasticizers studied. As shown in Fig. 9b, the more GTA added the less Ts of the polymers (i.e. EPO, SOL, and RL). The Ts reduction plateaued when the GTA concentration was higher than 30%. The effect of GTA concentration on the morphology and alignment of fibers is shown in Fig. 10 setting Ts as operation temperature (i.e. 130 °C for EPO, 150 °C for SOL and 160 °C for RL). The increase in GTA concentration facilitated the fiber formation. When GTA concentration was increased from 20% to 40%, highly aligned fibers were obtained. However, the produced fibers tended to stick on the packaging Fig. 6. Melting temperature or soft temperature (Tm or Ts) and melt operating temperature range (△T1) of polyethylene glycol (PEG), Eudragit® EPO (EPO), Soluplus® (SOL), Eudragit® RL PO (RL), Eudragit® RS PO (RS) and ethyl cellulose (EC).
159
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 7. Stereomicroscope photos of polymer fibers studied: polyethylene glycol (PEG), Eudragit® EPO (EPO), Soluplus® (SOL), Eudragit® RL PO (RL), Eudragit® RS PO (RS) and ethyl cellulose (EC).
shown). TNZ with poor drug fiber formability was also studied to understand the effect of drug loading on fiber formation. As shown in Fig. 13a, SOL and RL fibrous films with 20% TNZ had a random alignment, similar to their respective blank films. As shown in Fig. 13b, TNZ-loaded SOL fibrous film (TNZ-SOL film) showed immediate drug release characteristics, which was similar to the release profile of TNZ powder alone. TNZ is water soluble and accordingly, dissolution enhancement by forming fibrous films may not be necessary. On the other hand, sustained TNZ release was obtained by forming fibrous RL films (TNZ-RL film) (> 80% released by 12 h) although a high initial burst release (> 60% within 30 min) was observed. The fibrous film typically has a highly porous structure and high surface area. Consequently, watersoluble TNZ located at or near the surface of the fibers may be responsible for the high initial burst release.
materials during storage when GTA concentration was more than 20%. Therefore, it was determined that co-plasticizer(s) may be needed. PEG has been used as a plasticizer for various pharmaceutical polymers (Maru et al., 2011; Flösser et al., 2000; Laboulfie et al., 2013) and itself alone did not appear to be a good plasticizer for the polymers investigated (i.e. EPO, SOL, and RL) during the CMS process (data not shown). Considering that PEG is not sticky and it may work synergistically with GTA, the combination of GTA and PEG as co-plasticizers was also studied. As shown in Fig. 10, the combination of 20% GTA and 30% PEG resulted in better aligned and less sticky films compared to 20% GTA and 40% GTA, respectively. The effect of GTA concentration on the diameter and yield of EPO, SOL and RL fibers was further investigated. As shown in Fig. 11a, the diameter of polymer fibers significantly decreased with the increase in GTA concentration, especially for SOL fibers. SOL had a relatively high melt viscosity and the effect of GTA was more profound. As shown in Fig. 11b, the yield of EPO fibers was increased from 30% to 90% with the increase in GTA concentration from 0% to 40%. A similar trend was observed for SOL and RL.
3.3. Melt flow field simulation CMS is a complex and transient process, during which the materials melt and fiber formation via centrifuge force occur instantaneously. To simplify the simulation, the MRF model was adopted to couple the moving zone and static zone, and to simulate the steady-state of the fluid field. A typical two-dimensional plane projection model of CMS flow field was shown in Fig. 14a. The moving zone was consistent with 24 outlet channels, each with 3 spinneret orifices and a viscous fluid layer with a thickness of 1.0 mm. The area between viscous fluid layer and outer collecting barrel was set to be static zone. As shown in Fig. 14b, the moving zone was meshed into 3520 elements, while the static zone was mesh into 9878 elements when the outer collecting barrel had a diameter of 100 mm.
3.2.3. Influence of drug loading on fiber formation The fiber formability of drug-loaded fibrous films was also investigated using IND as a model drug. As shown in Fig. 12a, SOL and RL fibrous films with 20% (w/w) IND had a higher alignment compared to their respective polymer only films. As shown in Fig. 12b, the INDloaded SOL fibrous films (IND-SOL film) possessed immediate drug release characteristics (> 90% within 1 h), while the IND-loaded RL films were capable of sustained releasing drug (< 50% over 12 h). The addition of 30% PEG in the RL films appeared to accelerate drug release (> 70% released by 12 h), yet had no effect on the SOL films (data not
Fig. 8. Melt complex viscosity (η*) curves of (a) Eudragit® RL PO (RL), polyethylene glycol (PEG) and Soluplus® (SOL) during cooling; and (b) RL during a heating and cooling process. 160
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 9. (a) Influence of plasticizer type on the soft temperature (Ts) of Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL).The plasticizers studied were vitamin E acetate (VEA), triethyl citrate (TEC), glycerol triacetate (GTA) and polyethylene glycol (PEG). (b) Influence of GTA concentration on the Ts of EPO, SOL and RL.
while the contours in the outlet channel were dense and non-symmetrically distributed when partially enlarged (Fig. 15a, b, and c). The ultimate values for total pressure (0.473 MPa), turbulent flow energy (103 m2/s2) and velocity (27.9 m/s) were all appeared in the outlet
Assuming that the molten IND was viscous and incompressible at 160 °C (ρ = 1,107.4 kg/m3, η′ = 0.796 Pa·s), typical flow fields of the drug during CMS were simulated (Fig. 15). The total pressure, turbulent flow energy, and velocity were axisymmetric in the entire flow field,
Fig. 10. Influence of glycerol triacetate (GTA) concentration and the addition of polyethylene glycol (PEG) as co-plasticizers on the morphology and alignment of polymer fibers analyzed using stereomicroscope. The polymers studied were Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL). 161
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 11. Influence of glycerol triacetate (GTA) concentration on (a) diameter and (b) yield of Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL) fibers.
Fig. 12. (a)Stereomicroscope images of indomethacin (IND) fibrous films using Soluplus® (SOL) and Eudragit® RL PO (RL) as the carriers, and glycerol triacetate (GTA) or the combination of GTA and polyethylene glycol (PEG) as plasticizers. (b) In vitro dissolution profiles of IND-loaded fibrous films. IND powder was studied as a control.
Fig. 13. (a) Stereomicroscope images of Tinidazole (TNZ) fibrous films using Soluplus® (SOL) and Eudragit® RL PO (RL) as the polymer carriers. (b) In vitro dissolution profiles of TNZ-loaded fibrous films. TNZ powder was studied as a control.
162
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 14. Illustration of the two-dimensional plane projection of centrifugal melt spinning (CMS) flow field. (a) Geometric model and (b) mesh model.
Fig. 15. Illustration of the flow filed simulation of indomethacin (IND). (a) Total pressure contour; (b) Turbulent kinetic energy contour; (c) Velocity magnitude vectors; and (d) Partial pathlines colored by particle and rotation angle (θ) of single partial pathline.
Fig. 16. Illustration of the flow filed simulation of plasticized Eudragit® RL PO (RL). (a) Total pressure contour; (b) Turbulent kinetic energy contour; (c) Velocity magnitude vectors; and (d) Partial pathlines colored by particle and rotation angle (θ) of single partial pathline.
channel. There were no obvious backflow and turbulent flow observed (Fig. 15b and c). Moreover, the tangential velocity at the spinneret orifice was high (Fig. 15c), suggesting that the velocity vector was mainly caused by centrifugal force. The simulated partial pathlines visualized the CMS process (Fig. 15d). The fibrous films formed laterally between the collecting barrels, with non-parallel fiber arrangement due to the turbulent flow. The rotation angle (θ) of single partial pathline
from spinneret orifices to the collector was 340°. The assumption was that the molten RL with 20% GTA as a plasticizer was viscous and incompressible at 160 °C (ρ = 1140.9 kg/m3, η′ = 676.1 Pa·s), typical flow fields of the polymer during CMS were simulated (Fig. 16). The flow fields of the plasticized RL were axisymmetrically distributed similarly to that of IND mentioned above. As we expected, the contours distribution of RL were different from that of
163
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 17. Influence of (a) rotating speed, (b) inlet rate, (c) collecting radius, (d) melt density and (e) melt viscosity on the rotation angle and outlet velocity of IND fiber during the centrifugal melt spinning (CMS) process.
operation parameters and material properties. The results could be used to guide process and formulation optimization.
IND when the outlet channel was partially enlarged (Fig. 16a, b and c). Centrifugal force and channel morphology had a great impact on the plasticized RL with high viscosity than IND with low viscosity. The ultimate total pressure (47.8 MPa) and ultimate turbulent flow energy (45,100 m2/s2) of the plasticized RL were much higher than that of IND, while the ultimate velocity (24.0 m/s) appeared to be slightly lower. High static pressure, high turbulent flow energy and low kinetic energy associated with the polymer was due to its high viscosity and poor flow property. Similarly, the formation of lateral fibrous films between the collectors was due to the tangential velocity (Fig. 16c and d). Lastly, the θ of single partial pathline was reduced to be 167° (Fig. 16d). It can be explained by the high resistance and energy loss of highly viscous plasticized RL. Based on the IND simulation above, the effect of operation parameters (i.e. rotating speed, inlet rate, collecting radius) and material properties (i.e. density and viscosity) on the outlet velocity of products and sample collection position was evaluated (Fig. 17). The rotation angle increased with a higher rotational speed, a lower inlet rate or a reduced melt viscosity. The outlet velocity decreased with a lower inlet rate, a larger collection radius or an increased melt viscosity. The influence of melt density was relatively small. The outlet velocity of products and sample collection position could be adjusted by various
4. Conclusion A high throughput and one-step fabrication method via CMS technology was successfully developed to fabricate fibrous films. The produced films had high drug loading, highly aligned fiber orientation and film structure, as well as modulatable drug release characteristics. Maintaining suitable fluidity of molten materials was found to be critical for the formation of fibrous films. Furthermore, CFD method was capable of simulating melt flow fields and can potentially be used to assist in understanding fiber formation as well as optimization of CMS process parameters. Thermal and storage stabilities of fibrous films will be studied in details. In addition, a three-dimensional model with a coupled complex temperature and airflow field (Arne et al., 2011) will be established in future studies to improve the simulation accuracy. Unsteady flow analysis (Barrio et al., 2010) will also be conducted to further elucidate melt flow characteristics during the CMS process.
164
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
5. Declaration of interest statement
Laboulfie, F., et al., 2013. Effect of the plasticizer on permeability, mechanical resistance and thermal behaviour of composite coating films. Powder Technol. 238, 14–19. Li, X., et al., 2013. Electrospun polyvinyl-alcohol nanofibers as oral fast-dissolving delivery system of caffeine and riboflavin. Colloids Surf. B: Biointerfaces 103, 182–188. Li, C., et al., 2014. Higher quality quercetin sustained release ethyl cellulose nanofibers fabricated using a spinneret with a Teflon nozzle. Colloids Surf. B: Biointerfaces 114, 404–409. Loordhuswamy, A.M., et al., 2014. Fabrication of highly aligned fibrous scaffolds for tissue regeneration by centrifugal spinning technology. Mater. Sci. Eng. C 42, 799–807. Marano, S., et al., 2016. Development of micro-fibrous solid dispersions of poorly watersoluble drugs in sucrose using temperature-controlled centrifugal spinning. Eur. J. Pharm. Biopharm. 103, 84–94. Marano, S., et al., 2017. Microfibrous solid dispersions of poorly water-soluble drugs produced via centrifugal spinning: unexpected dissolution behavior on recrystallization. Mol. Pharm. 14, 1666–1680. Maru, S.M., et al., 2011. Characterization of thermal and rheological properties of zidovudine, lamivudine and plasticizer blends with ethyl cellulose to assess their suitability for hot melt extrusion. Eur. J. Pharm. Biopharm. 44, 471–478. Mencinger, J., et al., 2015. Numerical simulation of ligament-growth on a spinning wheel. Int. J. Multiphase Flow 77, 90–103. O’Haire, T., et al., 2016. Carr, centrifugal melt spinning of polyvinylpyrrolidone (PVP)/ triacontene copolymer fibres. J. Mater. Sci. 51, 7512–7522. Shimada, Y., et al., 2018. Decarboxylation of indomethacin induced by heat treatment. Int. J. Pharm. 545, 51–56. Wang, C., et al., 2017. Optimal design of multistage centrifugal pump based on the combined energy loss model and computational fluid dynamics. Appl. Energy 187, 10–26. Wang, Z., Suo, J., 2012. Preparation of a composite fibrous membrane loaded with mesalazine and metronidazole by interlaced electrospinning. Mol. Med. Rep. 5, 535–538. Yan, J., et al., 2013. Sustained-release multiple-component cellulose acetate nanofibers fabricated using a modified coaxial electrospinning process. J. Mater. Sci. 49, 538–547. Yan, E., et al., 2014. Biocompatible core-shell electrospun nanofibers as potential application for chemotherapy against ovary cancer. Mater. Sci. Eng. C 41, 217–223. Yang, S., et al., 2017a. Hydroxypropyltrimethyl ammonium chloride chitosan functionalized-PLGA electrospun fibrous membranes as antibacterial wound dressing: in vitro and in vivo evaluation. Polymers 9, 697. Yang, Y., et al., 2017b. Preparation and evaluation of metoprolol tartrate sustained release pellets using hot melt extrusion combined with hot melt coating. Drug Dev. Ind. Pharm. 43, 939–946. Yang, Y., et al., 2018. 3D printed tablets with internal scaffold structure using ethyl cellulose to achieve sustained ibuprofen release. Eur. J. Pharm. Biopharm. 115, 11–18. Zander, N.E., 2015. Formation of melt and solution spun polycaprolactone fibers by centrifugal spinning. J. Appl. Polym. Sci. 132 (41269), 1–9. Zhou, H., et al., 2018. Cotton-like micro- and nanoscale poly(lactic acid) nonwoven fibers fabricated by centrifugal melt-spinning for tissue engineering. RSC Adv. 8, 5166–5179.
None. Acknowledgments This work was supported by China Scholarship Council (CSC No. 201808330014), National Natural Science Foundation of China (No. 81402873), Public Welfare Technology Application Research Project of Zhejiang Province, China (No. 2015C33142). References Altamimi, M.A., Neau, S.H., 2018. A study to identify the contribution of Soluplus® component homopolymers to the solubilization of nifedipine and sulfamethoxazole using the melting point depression method. Powder Technol. 338, 576–585. Arne, W., et al., 2011. Fluid-fiber-interactions in rotational spinning process of glass wool production. J. Math. Ind. 1, 1–26. Barrio, R., et al., 2010. Numerical analysis of the unsteady flow in the near-tongue region in a volute-type centrifugal pump for different operating points. Comput. Fluids 39, 859–870. Dalton, P.D., et al., 2007. Electrospinning of polymer melts: phenomenological observations. Polymer 48, 6823–6833. Davidovich-Pinhas, M., et al., 2014. Physical structure and thermal behavior of ethylcellulose. Cellulose 21, 3243–3255. Flösser, A., et al., 2000. Variation of composition of an enteric formulation based on Kollicoat MAE 30 D. Drug Dev. Ind. Pharm. 26, 177–187. Fujimori, J., et al., 2002. Application of Eudragit RS to thermo-sensitive drug delivery systems: I. Thermo-sensitive drug release from acetaminophen matrix tablets consisting of Eudragit RS/PEG 400 blend polymers. Chem. Pharm. Bull. 50, 408–412. Gioumouxouzis, C.I., et al., 2018. A 3D printed bilayer oral solid dosage form combining metformin for prolonged and glimepiride for immediate drug delivery. Eur. J. Pharm. Biopharm. 120, 40–52. Grymonpré, W., et al., 2017. In-line monitoring of compaction properties on a rotary tablet press during tablet manufacturing of hot-melt extruded amorphous solid dispersions. Int. J. Pharm. 517, 348–358. Hassan, M.A., et al., 2016. Computational fluid dynamics simulations and experiments of meltblown fibrous media: new die designs to enhance fiber attenuation and filtration quality. Ind. Eng. Chem. Res. 55, 2049–2058. Illangakoon, U.E., et al., 2014. Fast dissolving paracetamol/caffeine nanofibers prepared by electrospinning. Int. J. Pharm. 477, 369–379. Jafarzadeh, B., et al., 2011. The flow simulation of a low-specific-speed high-speed centrifugal pump. Appl. Math. Modell. 35, 242–249. Khan, G., et al., 2016. Development, optimization and evaluation of tinidazole functionalized electrospun poly(ε-caprolactone) nanofiber membranes for the treatment of periodontitis. RSC Adv. 6, 100214–100229.
165
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Mechanistic study on rapid fabrication of fibrous films via centrifugal melt spinning ⁎
Yan Yanga,b, Nan Zhenga, Yanjun Zhoua, Weiguang Shana, , Jie Shenb, a b
T
⁎
College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, China College of Pharmacy, University of Rhode Island, Kingston, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Centrifugal melt spinning (CMS) Fibrous films Melt rheology Fiber formation mechanism Computational fluid dynamics Flow field analysis
Fibrous films have attracted considerable attention in the field of drug delivery and wound dressings owing to their porous structure and highly aligned fiber orientation. However, current fabrication methods such as electrospinning have certain limitations, including high voltage requirement and conductivity dependency. This has greatly hindered the product development and applications of fibrous films. The objective of the present study was to develop a high throughput and solventless fiber fabrication method via centrifugal melt spinning (CMS) technology. A mechanistic study on the rapid fabrication of drug-loaded fibrous films was conducted using different model drugs and polymers. It was observed that the formability, morphology, and yield of fibrous films were affected by melt rheological properties of film components, operation temperature, and plasticizers. Maintaining suitable fluidity of molten materials during the CMS process is critical for the fiber formation. The produced fibrous films had high drug loading, highly aligned orientation and modulatable drug dissolution characteristics. Finally, computational fluid dynamics (CFD) was used to simulate the melt flow fields during the CMS process. Pressure, turbulence, velocity, and partial pathlines were simulated to elucidate the influence of various operation parameters (i.e. rotating speed, inlet rate and collecting radius) and material properties (i.e. density and viscosity) on the outlet velocity of products and sample collection position. The present study demonstrated that CMS is a high throughput and cost-efficient fabrication method for drug-loaded fibrous films. CFD simulation can be used to assist in understanding fiber formation as well as optimization of CMS process parameters.
1. Introduction Fibrous films have gained widespread interest in the fields of pharmaceutics as drug delivery systems and bio-functional scaffolds, owing to their high surface area and drug loading capability, and highly porous structure and aligned fiber orientation. For example, fibrous film-based oral fast-dissolving films have been successfully produced (Illangakoon et al., 2014; Li et al., 2013). In addition, fibrous films have been used as functional wound dressings since they can absorb excess exudates, ensure sufficient gas and nutrient exchange and promote cell proliferation and migration (Yang et al., 2017a,b). Electrospinning is commonly used to produce fibrous films, in which a film component solution or molten material is (are) stretched into fibers via high voltage electrostatic force. Fibrous films based on pharmaceutical polymers such as polyvinylpyrrolidone (PVP) (Illangakoon et al., 2014), polyvinyl alcohol (PVA) (Li et al., 2013),
PVA/chitosan (Yan et al., 2014), poly(lactic-co-glycolic acid) (PLGA) (Wang and Suo, 2012), ethyl cellulose (EC) (Li et al., 2014) and cellulose acetate (Yan et al., 2013), have been produced using electrospinning. However, electrospinning technology has certain limitations, including high voltage (> 10 kV) requirement and conductivity dependency (Zander, 2015). An alternative approach to produce ultrafine fibers is centrifugal melt spinning (CMS), which employs centrifugal force to stretch the melt jet from the spinneret (Zander, 2015). CMS offers a solventless and high throughput fiber formation approach. It also enables fiber formation of polymers and/or drugs with poor or nonconductivity and hence broader applications. Most recently, CMS has been used to prepare sucrose-based microfibers containing poorly water-soluble drugs (i.e. olanzapine and piroxicam) (Marano et al., 2016). The drugs were found to be present in the amorphous state, and drug in vitro dissolution performance was significantly enhanced. Further stability study revealed that even aged fibers (freely recrystallized)
⁎ Corresponding authors at: College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, China, Tel.: +86-0571-88871075 (W. Shan) and College of Pharmacy, University of Rhode Island, Kingston, RI, United States Tel.: +1-401-874-5594 (J. Shen). E-mail addresses: [email protected] (W. Shan), [email protected] (J. Shen).
https://doi.org/10.1016/j.ijpharm.2019.02.005 Received 19 December 2018; Received in revised form 2 February 2019; Accepted 4 February 2019 Available online 12 February 2019 0378-5173/ © 2019 Elsevier B.V. All rights reserved.
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Nomenclature PVP PVA PLGA EC CMS CFD IBU TNZ MT NF IND PEG EPO RL RS SOL VEA
TEC GTA Tm Td TGA DSC η* η′ XRD MRF ρ Vin ω R △T1 △T2 Ts θ
polyvinylpyrrolidone polyvinyl alcohol polylactic-co-glycolic acid ethyl cellulose centrifugal melt spinning computational fluid dynamics ibuprofen tinidazole metoprolol tartrate nifedipine indomethacin polyethylene glycol Eudragit® EPO Eudragit® RL PO Eudragit® RS PO Soluplus® vitamin E acetate
triethyl citrate glycerol triacetate melting temperature thermal decomposition temperature thermal gravimetric analysis Differential scanning calorimetry complex viscosity dynamic viscosity X-ray diffraction multiple reference frame density inlet velocity rotational speed entrance radius melt operating temperature range fiber-forming temperature range soft temperature rotation angle
prepared using an in-house device, and various drugs and pharmaceutical polymers were tested for their fiber formability. The effects of film components, operation temperature and plasticizers on the formability, morphology, and yield of drug-loaded fibrous films were investigated. Melt rheology was employed to explore the thermoforming mechanism of film components (i.e. drugs with/without polymers). Lastly, CFD was implemented to simulate and assess the melt flow field based on twodimensional plane projection models.
retained dissolution enhancement capability (Marano et al., 2017). Despite of the fact that CMS has been shown potential in manufacturing drug-loaded fibers, the literature on the one-step fabrication of fibrous films via CMS has remained sparse. Commercial centrifugal spinning devices such as Cyclone Fiber Engine (FE 1.1M/S), Cyclone L1000 and cotton-candy machines can only be used to prepare fibers, not fibrous films and additional step(s) is(are) often needed to fabricate films from fibers, which is difficult to control and not cost effective. External collectors such as rotary drums (Zhou et al., 2018) and round bottom collectors (Loordhuswamy et al., 2014) have been used during the CMS process to collect fibrous scaffold. Till now, whether CMS can be used for one-step formation of drug-loaded fibrous films; and whether fibrous films can be used to achieve sustained drug release have not been studied or reported. Furthermore, fiber formation mechanism via the CMS technology has yet to be understood. Computational fluid dynamics (CFD) is a branch of fluid mechanics that uses numerical analysis and data structures to simulate the interactions of liquids with surfaces defined by boundary conditions. CFD has been used to simulate and analyze thermoforming techniques. For example, the airflow field of melt-blowing process was simulated using CFD (Hassan et al., 2016), and the results were used to improve die configurations to prepare melt blown fibers with reduced fiber diameter and pore size. In addition, CFD was used to simulate the jet trajectory of a spinning wheel atomizer (Mencingeret al., 2015) to investigate the influence of process variables on the ligament growth. Therefore, it was envisioned that CFD simulation might be a feasible way to visualize and analyze the fiber formation process during CMS and assist in the process and formulation optimization. The present study aims to conduct a mechanistic study on the rapid fabrication of drug-loaded fibrous films via CMS. Fibrous films were
2. Materials and methods 2.1. Materials Ibuprofen (IBU) (No. 20151224) was obtained from Zhenyuan Chemical Co., Ltd. (Hubei, China). Tinidazole (TNZ) (No. 164862), metoprolol tartrate (MT) (No. 131069) and nifedipine (NF) (No. 164862) were purchased from YuanCheng Pharmaceutical Co., Ltd. (Hubei, China). Indomethacin (IND) (No. 20160103) was supplied by XiYinHe Chemical Co., Ltd. (Hubei, China). Polyethylene glycol (PEG) (No. 60111, Mw = 6000) was purchased from XiLong Chemical Co., Ltd. (Guangxi, China). Eudragit® EPO (EPO) (No. G150131502), Eudragit® RL PO (RL) (No. G080936208) and Eudragit® RS PO (RS) (No. G111238239) were supplied by Evonik Rohm Co., Ltd. (Germany). Soluplus® (SOL) (No. 84414368EO) was provided by BASF (Germany). EC (Std. 20, No. PD403586) were obtained from Colorcon Coating Technology Co., Ltd. (Shanghai, China). Vitamin E acetate (VEA) (No. VA160404) was obtained from Healthful Biotechnology Co., Ltd. (Xian, China), triethyl citrate (TEC) (No. 150605) was obtained from FengyuanTushan Pharmaceutical Co., Ltd. (Anhui, China) and glycerol triacetate (GTA) (No. 20160623) was supplied by SuZhong Technology
Fig. 1. Illustration of the in-house centrifugal melt spinning (CMS) device. (a) Perspective view; and (b) top view. 156
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Greifensee, Switzerland) (Yang et al., 2018). IBU, TNZ, and MT were studied from 25 °C to 150 °C at a rate of 10 °C/min and IND and NF were tested from 25 °C to 200 °C at a rate of 10 °C/min. Thermal behavior of the model drugs was studied.
Table 1 Validated UV calibration curves for IND, NF and TNZ. Drugs
Wavelength
Equations
R
IND NF TNZ
320 nm 333 nm 317 nm
A = 0.0193C A = 0.0149C − 0.0048 A = 0.0362C − 0.0052
1.0000 0.9999 0.9999
2.4. Melt rheological analysis Melt rheology of the model drugs and polymers (i.e. PEG, SOL, and RL) were determined using an Anton Paar Physica Rheometer (MCR 302, Anton Paar, Austria) with a 25.0 mm parallel plate in an oscillation mode and a gap distance of 0.3 mm (Maru et al., 2011). Following the preliminary study, the frequency was set at an angular frequency of 10 rad/s with a strain amplitude of 0.5%. Complex viscosity (η*) and dynamic viscosity (η′) were recorded every 5 s. The equilibrium temperatures were 100 °C for IBU, 150 °C for MT, 180 °C for IND, 190 °C for NF, 100 °C for PEG, and 200 °C for both SOL and RL. The test was conducted at a cooling rate of 0.8 °C/min. The effect of heating and cooling process on the melt rheology of the film components was also investigated.
Chemical Co., Ltd. (Guangdong, China). 2.2. Preparation of fibrous films Fibrous films were prepared using an in-house CMS device (Fig. 1) composed of a melting chamber with a radius of 60 mm. There were 24 outlet sectoral channels (thickness 5.0 mm, 6.0 rad) and each with 3 × 3 spinneret orifices (thickness 0.5 mm, 0.5 rad) on the side wall of the melting chamber. The operation temperature of the melting chamber was precisely controlled via an infrared temperature sensor and a heating device, and adjusted for different film components. The film components (e.g. drugs or drug/polymer mixture) were mixed in a mortar for 10 min and fed into the melting chamber. The melting chamber was then rotated at a speed of 3000 rpm and products were collected in the concentric barrels (as shown in Fig. 1). Morphology and size of the collected products were determined using a stereomicroscope (ZY-HD1400, ZongyanWeiye, China) with S-EYE v1.1 measuring software.
XRD analysis of the products was performed in an X-ray diffractometer (D/max 2550/PC, Rigaku, Japan) with Cu Kα1 radiation (Shimada et al., 2018). Samples were scanned at 40 kV and 40 mA, in the 2θ range of 5–50° at 10°/min.
2.3. Thermal analysis
2.6. In vitro dissolution testing
Thermal analysis was conducted to determine melting temperature (Tm) and decomposition temperature (Td) of the model drugs (i.e. IBU, TNZ, MT, IND, and NF). To determine the Td, thermal gravimetric analysis (TGA) was conducted using a TG analyzer (Q5000, TA instrument, New Castle, USA) (Khan et al., 2016). Samples were heated from 25 °C to 400 °C at a rate of 10 °C/min under a nitrogen atmosphere, and the Td of the model drugs was determined. In addition, approximately 5 mg of the model drugs were weighed and analyzed using a differential scanning calorimeter (DSC1, Mettler-Toledo,
In vitro dissolution of the fabricated films was conducted using a paddle method (USP Apparatus II, 100 rpm, 37 ± 0.5 °C, 500 mL) (Yang et al., 2018). The release medium was PBS (pH 7.2) for IND, 0.5% SLS solution for NF, and deionized water for TNZ, respectively. At predetermined time intervals, 5 mL of dissolution media were withdrawn and replaced with fresh media. The release samples were filtered through 0.45-μm Millipore filters, and analyzed using a spectrophotometer (UV-2450, Shimadzu, Japan). The UV analytical methods were established and validated for the representative model drugs (i.e.
2.5. X-ray diffraction (XRD) analysis
Fig. 2. (a) Thermo gravimetric analysis (TGA) curves; (b) Differential scanning calorimetry (DSC) diagrams; and (c) Melting temperature (Tm) and melt operating temperature range (△T1) of ibuprofen (IBU), tinidazole (TNZ), metoprolol tartrate (MT), indomethacin (IND) and nifedipine (NF).
157
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
IND, NF, and TNZ) and the calibration curves are shown in Table 1. The cumulative drug release profiles of model drugs were determined. 2.7. Melt flow field simulation Melt flow field simulation was conducted using a two-dimensional plane projection model (Wang et al., 2017) that was meshed using Gambit 2.4 software. The flow domain was divided into a moving zone and a static zone, and Tri&Pave elements were generated using a smart mesh method. Steady simulation of the melt flow fields during the CMS process was conducted using Fluent 6.3 software. A multiple reference frame (MRF) model was used to simulate the axial flow and a standard κ-ε model was adopted to describe the turbulence. The PRESTO! algorithm was used to calculate pressure and the SIMPLE algorithm was used to couple pressure and velocity (Jafarzadeh et al., 2011). The density (ρ) of melts was determined using a Melt Indexer (XNR-400A, Jinhe Instrument, China), while the η′ of melts was determined based on the melt rheological analysis as described above (see 2.4). The inlet velocity (Vin) was calculated using the following equation:
vin = R × ω
Fig. 4. Influence of temperature on complex viscosity (η*) of ibuprofen (IBU), metoprolol tartrate (MT), indomethacin (IND) and nifedipine (NF).
of IND and NF didn’t change significantly until the temperature was 20–30 °C below the Tm of these two drugs (i.e. 162 °C for IND and 174 °C for NF). As a result, the molten IND and NF maintained fluidity to some extent and were stretched into fibers. In the case of MT, the η* value gradually increased when the temperature was below 126 °C (Tm of MT) and drastically surged when the temperature was below 108 °C, indicating that the fluidity of the molten MT was not sufficient for fiber formation. Most of the molten MT solidified into powders with a merely small part of short fiber formation. The effect of operating temperature on fiber formation was also investigated using the two model drugs (i.e. IND and NF) with good fiber formability. The fiber-forming temperature range (△T2) within which fibrous films can form via CMS was 162–186 °C for IND and 174–188 °C for NF, respectively. Unexpectedly, powder formation was observed at high operating temperature (beyond the △T2). Since the η* of the molten IND or NF was too low when the operating temperature was beyond the △T2, the melt jet may break into droplets due to centrifugal force and surface tension and hence powders were formed rather than fibers. These results demonstrated that melt rheology study is critical to determine suitable operating temperature during the CMS process.
(1)
where ω is the rotational speed and R is the entrance radius. 3. Results and discussion 3.1. Pure drug fibrous films 3.1.1. Fiber formability of different drugs The Td determined via TGA was found to be IBU < MT < NF < TNZ = IND (Fig. 2a), while DSC results showed that the Tm was IBU < MT < TNZ < IND < NF (Fig. 2b). Accordingly, the melt operating temperature range (△T1) between Tm and Td was NF < MT < IND < IBU < TNZ (Fig. 2c). The fiber formability of different model drugs was evaluated by setting their Tm as respective operating temperature. Following the fiber formation process, only IND and NF fibrous films (diameter: 5–20 μm) were obtained between the inner and outer collecting barrels (Fig. 3). In the case of IBU and TNZ, powders (size: 50–200 μm) were collected on the outer barrel, whereas a mixture of powder and short fibers were obtained for MT. Melt rheology study was conducted to understand the fiber formability of the model compounds studied. As shown in Fig. 4, all drugs (i.e. IBU, MT, IND, and NF) had a low η* (< 1 Pa·s) at high temperature (near or above its Tm) and the fluids behaved differently during cooling. The η* value of IBU drastically increased as soon as the temperature was below 70 °C. This indicated that the η* of the fluid may be too high to be stretched into fibers when the molten material left the spinneret orifices and hence IBU powders were obtained. In contrast, the η* value
3.1.2. Immediate release characteristics As shown in Fig. 5a, the cumulative drug released from the IND or NF fibrous films was over 90% at 10 min, and much higher dissolution rates and extent were obtained compared with the drugs alone. The XRD results in Fig. 5b showed that IND was in amorphous state and the crystallinity of NF reduced significantly in their respective fibrous films. Accordingly, the fast dissolution rates can be explained by not only the highly porous structure and high surface area of the fibrous films, but
Fig. 3. Products collected on the concentric barrels and their morphology observed using stereomicroscope. IBU: ibuprofen; TNZ: tinidazole; MT: metoprolol tartrate; IND: indomethacin; and NF: nifedipine. 158
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 5. (a) In vitro dissolution profiles and (b) XRD results of indomethacin (IND), IND fibrous film, nifedipine (NF) and NF fibrous film.
increased gradually when cooled from 200 °C to 100 °C. Similar to the fiber formation of pure drug discussed above (see 3.1.1), the moderate η* value facilitated the stretching of the molten polymers into fibers. In contrast, the η* value of PEG drastically increased and the melt jet of PEG solidified into powder without stretching. A high operating temperature was also tested for PEG. It was observed that the melt jet of PEG broke into droplets due to its low viscosity. Therefore, it seemed that only polymers possessing suitable fluidity (gradually increased η* value over a wide temperature range) were good candidates for fibrous film formation via CMS. In addition, the influence of a typical heating and cooling process on the melt rheology was studied using RL as the model polymer. As shown in Fig. 8b, viscosity hysteresis was observed during the heating and cooling process. The phenomenon was related to crystal melting and transformation of polymer, indicating that the internal structure of the polymer may have been changed during thermal operation (O’Haire et al., 2016).
also changes in the solid state of the model drugs. 3.2. Drug-loaded polymer fibrous films 3.2.1. Fiber formability of different polymers The fiber formability of commonly used pharmaceutical polymers was investigated in order to tailor drug release from fibrous films. The minimum operating temperature during CMS was set to be Tm for crystalline polymer (i.e. PEG) and soft temperature (Ts) for amorphous polymers (i.e. EPO, SOL, RL, RS, and EC), respectively. As shown in Fig. 6, the Tm or Ts was PEG < EPO < SOL < RL = RS < EC. Base on the reported Td of different polymers (RL = RS < EC < EPO = SOL < PEG) (Altamimi and Neau, 2018; Grymonpré et al., 2017; Gioumouxouzis et al., 2018; Fujimori et al., 2002; DavidovichPinhas et al., 2014), the △T1, which was the temperature range from Tm or Ts to Td, was RL = RS < EC < SOL < EPO < PEG. The fiber formability of different polymers was evaluated by setting their Tm or Ts as the operating temperature (Fig. 7). Among the six polymers investigated, PEG formed powders with a diameter of 50–200 μm during the CMS process. SOL and EPO formed fibers with diameters of approximately 260 μm and 130 μm, respectively. The larger diameter of SOL might be due to its higher melt viscosity. In spite of a narrow △T1 of 10 °C (Fig. 6), RL and RS also formed uniform fibers with a diameter of approximately 190 μm. Considering that the film components only exposed to high temperature at a very short period of time, production of fibrous films via the CMS process with the instantaneous fiber formation is feasible for thermosensitive materials such as RL (Marano et al., 2016). In addition, EC fiber with a diameter of approximately 20 μm was obtained. Small fiber diameter may has resulted from its low melt viscosity at Ts of 170 °C. The slightly yellow appearance of EC fibers suggested that EC might be oxidized in air at high temperature during the CMS process. Melt rheology study was conducted to explain the different behaviors of the molten polymers (i.e. RL, SOL, and PEG) after leaving the spinneret orifice. As shown in Fig. 8a, the η* value of RL and SOL
3.2.2. Influence of plasticizer type and concentration on fiber formation It was reported that addition of plasticizers such as DBS and TEC can lower the melt processing temperature (Gioumouxouzis et al., 2018; Yang et al., 2017a,b) and refine the fibers by lowering the viscosity of polymer melts (Dalton et al., 2007). The effect of plasticizer type and concentration on Ts of EPO, SOL and RL was evaluated. As shown in Fig. 9a, the addition plasticizer reduced the Ts of polymers, and GTA (20%, w/v) was the best plasticizer among the plasticizers studied. As shown in Fig. 9b, the more GTA added the less Ts of the polymers (i.e. EPO, SOL, and RL). The Ts reduction plateaued when the GTA concentration was higher than 30%. The effect of GTA concentration on the morphology and alignment of fibers is shown in Fig. 10 setting Ts as operation temperature (i.e. 130 °C for EPO, 150 °C for SOL and 160 °C for RL). The increase in GTA concentration facilitated the fiber formation. When GTA concentration was increased from 20% to 40%, highly aligned fibers were obtained. However, the produced fibers tended to stick on the packaging Fig. 6. Melting temperature or soft temperature (Tm or Ts) and melt operating temperature range (△T1) of polyethylene glycol (PEG), Eudragit® EPO (EPO), Soluplus® (SOL), Eudragit® RL PO (RL), Eudragit® RS PO (RS) and ethyl cellulose (EC).
159
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 7. Stereomicroscope photos of polymer fibers studied: polyethylene glycol (PEG), Eudragit® EPO (EPO), Soluplus® (SOL), Eudragit® RL PO (RL), Eudragit® RS PO (RS) and ethyl cellulose (EC).
shown). TNZ with poor drug fiber formability was also studied to understand the effect of drug loading on fiber formation. As shown in Fig. 13a, SOL and RL fibrous films with 20% TNZ had a random alignment, similar to their respective blank films. As shown in Fig. 13b, TNZ-loaded SOL fibrous film (TNZ-SOL film) showed immediate drug release characteristics, which was similar to the release profile of TNZ powder alone. TNZ is water soluble and accordingly, dissolution enhancement by forming fibrous films may not be necessary. On the other hand, sustained TNZ release was obtained by forming fibrous RL films (TNZ-RL film) (> 80% released by 12 h) although a high initial burst release (> 60% within 30 min) was observed. The fibrous film typically has a highly porous structure and high surface area. Consequently, watersoluble TNZ located at or near the surface of the fibers may be responsible for the high initial burst release.
materials during storage when GTA concentration was more than 20%. Therefore, it was determined that co-plasticizer(s) may be needed. PEG has been used as a plasticizer for various pharmaceutical polymers (Maru et al., 2011; Flösser et al., 2000; Laboulfie et al., 2013) and itself alone did not appear to be a good plasticizer for the polymers investigated (i.e. EPO, SOL, and RL) during the CMS process (data not shown). Considering that PEG is not sticky and it may work synergistically with GTA, the combination of GTA and PEG as co-plasticizers was also studied. As shown in Fig. 10, the combination of 20% GTA and 30% PEG resulted in better aligned and less sticky films compared to 20% GTA and 40% GTA, respectively. The effect of GTA concentration on the diameter and yield of EPO, SOL and RL fibers was further investigated. As shown in Fig. 11a, the diameter of polymer fibers significantly decreased with the increase in GTA concentration, especially for SOL fibers. SOL had a relatively high melt viscosity and the effect of GTA was more profound. As shown in Fig. 11b, the yield of EPO fibers was increased from 30% to 90% with the increase in GTA concentration from 0% to 40%. A similar trend was observed for SOL and RL.
3.3. Melt flow field simulation CMS is a complex and transient process, during which the materials melt and fiber formation via centrifuge force occur instantaneously. To simplify the simulation, the MRF model was adopted to couple the moving zone and static zone, and to simulate the steady-state of the fluid field. A typical two-dimensional plane projection model of CMS flow field was shown in Fig. 14a. The moving zone was consistent with 24 outlet channels, each with 3 spinneret orifices and a viscous fluid layer with a thickness of 1.0 mm. The area between viscous fluid layer and outer collecting barrel was set to be static zone. As shown in Fig. 14b, the moving zone was meshed into 3520 elements, while the static zone was mesh into 9878 elements when the outer collecting barrel had a diameter of 100 mm.
3.2.3. Influence of drug loading on fiber formation The fiber formability of drug-loaded fibrous films was also investigated using IND as a model drug. As shown in Fig. 12a, SOL and RL fibrous films with 20% (w/w) IND had a higher alignment compared to their respective polymer only films. As shown in Fig. 12b, the INDloaded SOL fibrous films (IND-SOL film) possessed immediate drug release characteristics (> 90% within 1 h), while the IND-loaded RL films were capable of sustained releasing drug (< 50% over 12 h). The addition of 30% PEG in the RL films appeared to accelerate drug release (> 70% released by 12 h), yet had no effect on the SOL films (data not
Fig. 8. Melt complex viscosity (η*) curves of (a) Eudragit® RL PO (RL), polyethylene glycol (PEG) and Soluplus® (SOL) during cooling; and (b) RL during a heating and cooling process. 160
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 9. (a) Influence of plasticizer type on the soft temperature (Ts) of Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL).The plasticizers studied were vitamin E acetate (VEA), triethyl citrate (TEC), glycerol triacetate (GTA) and polyethylene glycol (PEG). (b) Influence of GTA concentration on the Ts of EPO, SOL and RL.
while the contours in the outlet channel were dense and non-symmetrically distributed when partially enlarged (Fig. 15a, b, and c). The ultimate values for total pressure (0.473 MPa), turbulent flow energy (103 m2/s2) and velocity (27.9 m/s) were all appeared in the outlet
Assuming that the molten IND was viscous and incompressible at 160 °C (ρ = 1,107.4 kg/m3, η′ = 0.796 Pa·s), typical flow fields of the drug during CMS were simulated (Fig. 15). The total pressure, turbulent flow energy, and velocity were axisymmetric in the entire flow field,
Fig. 10. Influence of glycerol triacetate (GTA) concentration and the addition of polyethylene glycol (PEG) as co-plasticizers on the morphology and alignment of polymer fibers analyzed using stereomicroscope. The polymers studied were Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL). 161
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 11. Influence of glycerol triacetate (GTA) concentration on (a) diameter and (b) yield of Eudragit® EPO (EPO), Soluplus® (SOL) and Eudragit® RL PO (RL) fibers.
Fig. 12. (a)Stereomicroscope images of indomethacin (IND) fibrous films using Soluplus® (SOL) and Eudragit® RL PO (RL) as the carriers, and glycerol triacetate (GTA) or the combination of GTA and polyethylene glycol (PEG) as plasticizers. (b) In vitro dissolution profiles of IND-loaded fibrous films. IND powder was studied as a control.
Fig. 13. (a) Stereomicroscope images of Tinidazole (TNZ) fibrous films using Soluplus® (SOL) and Eudragit® RL PO (RL) as the polymer carriers. (b) In vitro dissolution profiles of TNZ-loaded fibrous films. TNZ powder was studied as a control.
162
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 14. Illustration of the two-dimensional plane projection of centrifugal melt spinning (CMS) flow field. (a) Geometric model and (b) mesh model.
Fig. 15. Illustration of the flow filed simulation of indomethacin (IND). (a) Total pressure contour; (b) Turbulent kinetic energy contour; (c) Velocity magnitude vectors; and (d) Partial pathlines colored by particle and rotation angle (θ) of single partial pathline.
Fig. 16. Illustration of the flow filed simulation of plasticized Eudragit® RL PO (RL). (a) Total pressure contour; (b) Turbulent kinetic energy contour; (c) Velocity magnitude vectors; and (d) Partial pathlines colored by particle and rotation angle (θ) of single partial pathline.
channel. There were no obvious backflow and turbulent flow observed (Fig. 15b and c). Moreover, the tangential velocity at the spinneret orifice was high (Fig. 15c), suggesting that the velocity vector was mainly caused by centrifugal force. The simulated partial pathlines visualized the CMS process (Fig. 15d). The fibrous films formed laterally between the collecting barrels, with non-parallel fiber arrangement due to the turbulent flow. The rotation angle (θ) of single partial pathline
from spinneret orifices to the collector was 340°. The assumption was that the molten RL with 20% GTA as a plasticizer was viscous and incompressible at 160 °C (ρ = 1140.9 kg/m3, η′ = 676.1 Pa·s), typical flow fields of the polymer during CMS were simulated (Fig. 16). The flow fields of the plasticized RL were axisymmetrically distributed similarly to that of IND mentioned above. As we expected, the contours distribution of RL were different from that of
163
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
Fig. 17. Influence of (a) rotating speed, (b) inlet rate, (c) collecting radius, (d) melt density and (e) melt viscosity on the rotation angle and outlet velocity of IND fiber during the centrifugal melt spinning (CMS) process.
operation parameters and material properties. The results could be used to guide process and formulation optimization.
IND when the outlet channel was partially enlarged (Fig. 16a, b and c). Centrifugal force and channel morphology had a great impact on the plasticized RL with high viscosity than IND with low viscosity. The ultimate total pressure (47.8 MPa) and ultimate turbulent flow energy (45,100 m2/s2) of the plasticized RL were much higher than that of IND, while the ultimate velocity (24.0 m/s) appeared to be slightly lower. High static pressure, high turbulent flow energy and low kinetic energy associated with the polymer was due to its high viscosity and poor flow property. Similarly, the formation of lateral fibrous films between the collectors was due to the tangential velocity (Fig. 16c and d). Lastly, the θ of single partial pathline was reduced to be 167° (Fig. 16d). It can be explained by the high resistance and energy loss of highly viscous plasticized RL. Based on the IND simulation above, the effect of operation parameters (i.e. rotating speed, inlet rate, collecting radius) and material properties (i.e. density and viscosity) on the outlet velocity of products and sample collection position was evaluated (Fig. 17). The rotation angle increased with a higher rotational speed, a lower inlet rate or a reduced melt viscosity. The outlet velocity decreased with a lower inlet rate, a larger collection radius or an increased melt viscosity. The influence of melt density was relatively small. The outlet velocity of products and sample collection position could be adjusted by various
4. Conclusion A high throughput and one-step fabrication method via CMS technology was successfully developed to fabricate fibrous films. The produced films had high drug loading, highly aligned fiber orientation and film structure, as well as modulatable drug release characteristics. Maintaining suitable fluidity of molten materials was found to be critical for the formation of fibrous films. Furthermore, CFD method was capable of simulating melt flow fields and can potentially be used to assist in understanding fiber formation as well as optimization of CMS process parameters. Thermal and storage stabilities of fibrous films will be studied in details. In addition, a three-dimensional model with a coupled complex temperature and airflow field (Arne et al., 2011) will be established in future studies to improve the simulation accuracy. Unsteady flow analysis (Barrio et al., 2010) will also be conducted to further elucidate melt flow characteristics during the CMS process.
164
International Journal of Pharmaceutics 560 (2019) 155–165
Y. Yang, et al.
5. Declaration of interest statement
Laboulfie, F., et al., 2013. Effect of the plasticizer on permeability, mechanical resistance and thermal behaviour of composite coating films. Powder Technol. 238, 14–19. Li, X., et al., 2013. Electrospun polyvinyl-alcohol nanofibers as oral fast-dissolving delivery system of caffeine and riboflavin. Colloids Surf. B: Biointerfaces 103, 182–188. Li, C., et al., 2014. Higher quality quercetin sustained release ethyl cellulose nanofibers fabricated using a spinneret with a Teflon nozzle. Colloids Surf. B: Biointerfaces 114, 404–409. Loordhuswamy, A.M., et al., 2014. Fabrication of highly aligned fibrous scaffolds for tissue regeneration by centrifugal spinning technology. Mater. Sci. Eng. C 42, 799–807. Marano, S., et al., 2016. Development of micro-fibrous solid dispersions of poorly watersoluble drugs in sucrose using temperature-controlled centrifugal spinning. Eur. J. Pharm. Biopharm. 103, 84–94. Marano, S., et al., 2017. Microfibrous solid dispersions of poorly water-soluble drugs produced via centrifugal spinning: unexpected dissolution behavior on recrystallization. Mol. Pharm. 14, 1666–1680. Maru, S.M., et al., 2011. Characterization of thermal and rheological properties of zidovudine, lamivudine and plasticizer blends with ethyl cellulose to assess their suitability for hot melt extrusion. Eur. J. Pharm. Biopharm. 44, 471–478. Mencinger, J., et al., 2015. Numerical simulation of ligament-growth on a spinning wheel. Int. J. Multiphase Flow 77, 90–103. O’Haire, T., et al., 2016. Carr, centrifugal melt spinning of polyvinylpyrrolidone (PVP)/ triacontene copolymer fibres. J. Mater. Sci. 51, 7512–7522. Shimada, Y., et al., 2018. Decarboxylation of indomethacin induced by heat treatment. Int. J. Pharm. 545, 51–56. Wang, C., et al., 2017. Optimal design of multistage centrifugal pump based on the combined energy loss model and computational fluid dynamics. Appl. Energy 187, 10–26. Wang, Z., Suo, J., 2012. Preparation of a composite fibrous membrane loaded with mesalazine and metronidazole by interlaced electrospinning. Mol. Med. Rep. 5, 535–538. Yan, J., et al., 2013. Sustained-release multiple-component cellulose acetate nanofibers fabricated using a modified coaxial electrospinning process. J. Mater. Sci. 49, 538–547. Yan, E., et al., 2014. Biocompatible core-shell electrospun nanofibers as potential application for chemotherapy against ovary cancer. Mater. Sci. Eng. C 41, 217–223. Yang, S., et al., 2017a. Hydroxypropyltrimethyl ammonium chloride chitosan functionalized-PLGA electrospun fibrous membranes as antibacterial wound dressing: in vitro and in vivo evaluation. Polymers 9, 697. Yang, Y., et al., 2017b. Preparation and evaluation of metoprolol tartrate sustained release pellets using hot melt extrusion combined with hot melt coating. Drug Dev. Ind. Pharm. 43, 939–946. Yang, Y., et al., 2018. 3D printed tablets with internal scaffold structure using ethyl cellulose to achieve sustained ibuprofen release. Eur. J. Pharm. Biopharm. 115, 11–18. Zander, N.E., 2015. Formation of melt and solution spun polycaprolactone fibers by centrifugal spinning. J. Appl. Polym. Sci. 132 (41269), 1–9. Zhou, H., et al., 2018. Cotton-like micro- and nanoscale poly(lactic acid) nonwoven fibers fabricated by centrifugal melt-spinning for tissue engineering. RSC Adv. 8, 5166–5179.
None. Acknowledgments This work was supported by China Scholarship Council (CSC No. 201808330014), National Natural Science Foundation of China (No. 81402873), Public Welfare Technology Application Research Project of Zhejiang Province, China (No. 2015C33142). References Altamimi, M.A., Neau, S.H., 2018. A study to identify the contribution of Soluplus® component homopolymers to the solubilization of nifedipine and sulfamethoxazole using the melting point depression method. Powder Technol. 338, 576–585. Arne, W., et al., 2011. Fluid-fiber-interactions in rotational spinning process of glass wool production. J. Math. Ind. 1, 1–26. Barrio, R., et al., 2010. Numerical analysis of the unsteady flow in the near-tongue region in a volute-type centrifugal pump for different operating points. Comput. Fluids 39, 859–870. Dalton, P.D., et al., 2007. Electrospinning of polymer melts: phenomenological observations. Polymer 48, 6823–6833. Davidovich-Pinhas, M., et al., 2014. Physical structure and thermal behavior of ethylcellulose. Cellulose 21, 3243–3255. Flösser, A., et al., 2000. Variation of composition of an enteric formulation based on Kollicoat MAE 30 D. Drug Dev. Ind. Pharm. 26, 177–187. Fujimori, J., et al., 2002. Application of Eudragit RS to thermo-sensitive drug delivery systems: I. Thermo-sensitive drug release from acetaminophen matrix tablets consisting of Eudragit RS/PEG 400 blend polymers. Chem. Pharm. Bull. 50, 408–412. Gioumouxouzis, C.I., et al., 2018. A 3D printed bilayer oral solid dosage form combining metformin for prolonged and glimepiride for immediate drug delivery. Eur. J. Pharm. Biopharm. 120, 40–52. Grymonpré, W., et al., 2017. In-line monitoring of compaction properties on a rotary tablet press during tablet manufacturing of hot-melt extruded amorphous solid dispersions. Int. J. Pharm. 517, 348–358. Hassan, M.A., et al., 2016. Computational fluid dynamics simulations and experiments of meltblown fibrous media: new die designs to enhance fiber attenuation and filtration quality. Ind. Eng. Chem. Res. 55, 2049–2058. Illangakoon, U.E., et al., 2014. Fast dissolving paracetamol/caffeine nanofibers prepared by electrospinning. Int. J. Pharm. 477, 369–379. Jafarzadeh, B., et al., 2011. The flow simulation of a low-specific-speed high-speed centrifugal pump. Appl. Math. Modell. 35, 242–249. Khan, G., et al., 2016. Development, optimization and evaluation of tinidazole functionalized electrospun poly(ε-caprolactone) nanofiber membranes for the treatment of periodontitis. RSC Adv. 6, 100214–100229.
165