dielectric mijloc intreelectrozi

8/9/2019 Dielectric Mijloc Intreelectrozi

1/11

A unique device for controlled electrospinning

S.B. Mitchell, J.E. SandersDepartment of Bioengineering, University of Washington, Seattle, Washington 98195

Received 10 January 2005; revised 3 August 2005; accepted 17 November 2005Published online 7 April 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30673

Abstract: The purpose of this research was to develop asystem for controlled electrospinning of fibro-porous scaf-folds for tissue engineering applications and to use thissystem to assess mesh architecture sensitivity to manufac-turing parameters. The intent was to achieve scaffolds withwell-controlled fiber diameters and inter-fiber spacing. To

accomplish these objectives, a custom, closed-loop con-trolled, electrospinning system was built. The system wasunique in that it had a collection surface that was indepen-dent of the electrodes. The system allowed independentmanipulation and analysis of a number of manufacturingparameters: distance between the electrodes, distance fromthe nozzle to the collection surface, applied voltage, temper-ature of the melt, collection surface dielectric strength, andcollection surface area. Morphological analysis of fabricatedmeshes showed that all test parameters significantly affectedfiber diameter and inter-fiber spacing. Further, contrary towhat is generally accepted in the electrospinning literature,

voltage and temperature (inversely related to viscosity)were not the most significant parameters. Features of thecollection surface, including dielectric strength and surfacearea, were more significant. This dominance is, in part, areflection of the unique electrospinning system used. Thecollection surface, which was not connected to either of the

electrodes, substantially altered the electric field between theelectrodes. Using the developed controlled electrospinningsystem, thermoplastic polyurethane meshes with fiber diam-eters ranging from 5 to 18 m with variability less than 1.8%were made; inter-fiber spacing ranged from 4 to 90 m withvariability less than 20.2%. The system has potential use in

biomedical applications where meshes with controlled fiberdiameter and inter-fiber spacing are of interest. 2006 WileyPeriodicals, Inc. J Biomed Mater Res 78A: 110120, 2006

Key words: electrospinning; encapsulation; fibro-porous;tissue engineering; biomaterial; polymer

INTRODUCTION

In biomedical implants, material architecture influ-ences the local tissue response. Materials with smallfibers and small pore spacing may help to reduce oreliminate the fibrous capsule that typically surroundsimplanted medical devices. Results from several stud-ies suggest that fiber diameters less than 15 m andinter-fiber or pore spacing less than 60 m are favor-able.16 It is difficult, however, to fabricate fibro-po-rous meshes with small fiber diameters and spacing,

without inducing flocculation (aggregation of fibers).Achieving consistent fiber diameters can also be prob-lematic.

Electrospinning is a manufacturing means for mak-ing nonwoven, small-diameter, fibro-porous materi-als. It has been used for various applications including

protective clothing,7 filtration,8 drug delivery devices,9

implant interfaces,10 and scaffolds for biomaterial de-vices or tissue substitutes.1113 Electrospinning in-volves applying a high voltage between a capillarycontaining a polymer solution (or melt) and a collec-tion surface. When a drop of the polymer forms at theend of the capillary and the high voltage is applied, acharge is induced on the surface of the droplet. Whenthe mutual charge repulsion exceeds the surface ten-sion in the droplet, the droplet deforms into a conicalshape known as a Taylor cone. Once a critical voltage

is exceeded, a charged jet ejects from the apex of thecone and deposits on the collection surface.1416

Modern day electrospinning is usually performedusing a syringe filled with solvated or heated polymer.The anode is connected to a hypodermic needle on thesyringe, and the cathode is connected to a collectionsurface. When the high voltage is applied, the Taylorcone forms, the polymer ejects from the cone apex,and the fibers collect on the grounded surface. Typicalvariations on this theme included the following: usinga syringe pump to control the polymer flow rate17;using a rotating mandrel or disk to collect the fi-bers18,19; manipulating the electric field20; and chang-ing the collection surface material or shape.8

Correspondence to: J.E. Sanders; e-mail: [email protected]

Contract grant sponsor: National Science Foundation En-gineering Research Center Program; contract grant number:EEC-9529161

2006 Wiley Periodicals, Inc.


2/11

Considerable research has been conducted, investi-gating the electrospinning phenomenon. Mathemati-cal models have been developed to examine the Taylorcone formation,1415,21 polymer jet emergence,1416 as-ymptotic decay,22 stabilities,17,23,24 and bending14,24 ofthe jet. The dependence of surface features such as

beading and porosity on polymer, polymer concen-trations, and process variables is also well investi-gated,12,2529 These studies are useful toward betterunderstanding of how electrospinning is achieved andthe material surface characteristics are generated.

It is reasonably well accepted in the literature thatthe main parameters affecting mesh architectural fea-tures of interest in biomedical applications, fiber di-ameter and inter-fiber spacing, are the applied voltageand the polymer viscosity,17,20,23,25,3033 Polymer vis-cosity is typically controlled by adjusting the temper-ature of the melt. For most polymers, viscosity andtemperature are inversely related,34 However, thestrong dependence of fiber diameter and inter-fiberspacing on voltage and polymer viscosity (inverselyrelated to temperature) is not quantitatively well sup-ported. Further, in the few quantitative studies pub-lished, there are inconsistencies as to how fiber diam-eter and inter-fiber spacing, and features related tothem, are affected by applied voltage and polymerviscosity. Increasing the voltage has been shown toincrease jet diameter35 and to increase polymer flowrate16,17 and bending instability30,32,36; the latter twofeatures are correlated with fiber diameter. However,in other investigations, increasing the voltage was

shown to decrease the jet diameter,10,16,21,30,37,38

de-crease the fiber diameter,10,25 and enhance the align-ment of the fibers.8 Furthermore, Noris et al.27 notedthat fiber diameter was independent of polymer solu-tion concentration (related to viscosity), whereas oth-ers indicated that concentration had a major influenceon fiber diameter.32,33,35,39

We expect that it is the lack of control and consis-tency of other electrospinning manufacturing param-eters in previous reports that explain the inconsisten-cies in the results. Part of the difficulty is that thepractice of grounding the collection surface to the

lower electrode does not allow the manufacturing fea-tures to be independently controlled. An electrospin-ning system with the collection surface independent ofthe electrodes is needed to systematically determinethe influence of manufacturing features. Features ex-pected of relevance would be those that induce thegreatest change in the electric field magnitude anddirection; these include distance between the elec-trodes, distance from the nozzle to the collection sur-face, collection surface dielectric strength, and collec-tion surface area. An important purpose of thisresearch was to conduct a sensitivity analysis of fiberdiameter and inter-fiber spacing to applied voltage,

polymer temperature, and other manufacturing pa-

rameters. This information is potentially useful to-ward design efforts to engineer meshes of specificarchitectures for particular applications and towardthe design of a feedback-controlled electrospinningmanufacturing system. It was of further interest toestablish possible consistencies in fiber diameter and

inter-fiber spacing, using a well-controlled electro-spinning device.To conduct such sensitivity analysis, first, a system

that allows control over all of these manufacturingparameters simultaneously was needed. Systems pre-viously described in the literature typically allowedfor precise control of the applied voltage, with onlygross estimates of polymer feed rate, viscosity, collec-tion distance, or other features. Further, to eliminatethe influence of solvent evaporation on the results, asystem that spun from a polymer melt was necessary.

The objective of this research was threefold: (1) todevelop an electrospinning system that allowedclosed-loop control of the following parameters: dis-tance between the electrodes, distance from the nozzleto the collection surface, applied voltage, temperatureof the melt (inversely related to the viscosity), collec-tion surface dielectric strength, and collection surfacearea; (2) to conduct a sensitivity analysis of fiber di-ameter and inter-fiber spacing to these electrospinningmanufacturing parameters; (3) to determine the con-sistency in inter-fiber diameter and fiber spacing thatcan be achieved with this controlled electrospinningsystem.

MATERIALS AND METHODS

Instrumentation

Mechanical system

A custom electrospinning system that allowed closed-loop control of the features of interest, listed above, wasdeveloped. The system allowed the electrodes to be posi-tioned independent of the collection surface. In addition to

avoiding the influences of grounding (or charging) the col-lection surface, this configuration enabled the exploration ofthe influence of collection surface characteristics on mesharchitecture and fiber morphology. In the future, indepen-dence of the collection surface and electrodes, through po-sition control of the collection surface during electrospin-ning, may allow the development of variable architecturewith complex, three-dimensional shapes.

The primary components of the electrospinning apparatuswere the polymer chamber, electrodes, enclosure, collectionsurface, and x-y-z translation stage (Fig. 1). The polymerchamber held 16 cm3 of polymer melt at a controlled tem-perature and pressure. The chamber was heated via bandheaters (model NHW00158, Tempco, Wood Dale, IL) placedaround the outer surface of the chamber. Temperature was

A UNIQUE DEVICE FOR CONTROLLED ELECTROSPINNING 111

Journal of Biomedical Materials Research Part ADOI 10.1002/jbm.a


3/11

monitored using platinum resistance temperature transduc-ers (RTD) (model 1PT100KN815, Omega, Stamford, CT) em-

bedded in the bottom of the chamber. A pressure controlvalve (IP413020, Omega) and a vacuum control valve

(PV1045V, Omega) were connected to ports on the top ofthe chamber, and pressure and vacuum sensors (PX142030D5V, PX141015V5V, Omega) were located in the lines

between the control and chamber ports. Temperature, pres-sure, and vacuum were controlled via closed-loop control-lers as described later. The grounded electrode was posi-tioned within the polymer melt chamber, extending outthrough an upper port. Though typically the positive elec-trode is positioned within the melt, this practice was notfollowed here because of the risk of arcing to the RTDs. Astainless steel nozzle with a length to diameter ratio of 4.516

was positioned on the bottom of the chamber, so that thepolymer emerging from the chamber entered the enclosure.The polymer chamber was mounted directly on top of thepolycarbonate enclosure, via high-temperature thermoplas-tic mounting brackets. The mounting brackets were de-signed in such a way that only the nozzle was exposed to theinterior of the enclosure. Iris ports on each side preventedairflow from affecting the electrospinning process, yet al-lowed easy entry into the enclosure. When assembled, thepolymer chamber was sealed, except for the inlets for pres-sure and vacuum control and the nozzle.

The positive electrode was mounted to a Delrin carriagewithin the enclosure. The carriage was connected to a0.8-mm lead screw so that the electrode distance from thenozzle could be easily adjusted. A high-voltage power sup-ply (RP-50100, Del, Valhalla, NY), connected between the

positive and ground electrodes, provided a controllable po-tential difference from 0 to 50 kV.

The collection surface was mounted to an x-y-z stage(404150XR, Parker Daedal, Harrison City, PA) via a polycar-

bonate arm that protruded into the enclosure through a portin the rear. An accordion boot formed a seal between thearm and the enclosure. All components within the enclosurewere constructed from dielectrics, with the exception of thenozzle and the positive electrode.

Controller

A Labview (National Instruments, Austin TX) virtual in-strument (VI) was used to control the temperature, voltage,

vacuum/pressure behind the melt, and stage position. Fortemperature measurement, signals from the RTDs were con-verted to voltages with signal conditioning modules (OM5-IP-600-C, Omega, Stamford, CT), digitized with a multifunc-

tion data acquisition board (PCI-MIO-16XE-10, NationalInstruments, Austin TX), and monitored on a PC (450 MHzOptiplex GX1, Dell, Round Rock, TX). To control the dura-tion of heating so as to achieve set point temperatures,pulse-width-modulated digital outputs from the controllerwere sent to solid-state control-output modules (ACO5-C,Omega, Stamford, CT). The duty cycle was determined us-ing the average temperature as a percentage of the set point.Lower percentages increased the duty cycle.

Vacuum and pressure were controlled with a proportionalclosed-loop feedback control scheme. Signals from the vac-uum sensor and pressure sensor were digitized and moni-tored on the PC. Signals sent to the vacuum control valveand pressure controller were proportional to the differences

between the set points and their respective instantaneousvalues. A greater difference increased the time the valveswere open.

Components from a library of Labview virtual instru-ments (Motion Toolbox, Parker Compumotor, RohnertPark, CA) were embedded in the VI for motion control of thex-y-zstage. Control command inputs from the VI were sentto an AT6400 four-axis controller (Compumotor) that con-verted the data for use in a step indexer (Zeta4, Compumo-tor). On initialization, all axes were positioned with respectto a known position. Monitoring the position with theAT6400 allowed for pseudo-feedback control of axis posi-tion.

Calibration

To calibrate the temperature transducers, the polymerchamber was filled with water or olive oil, and then, heatedusing the band heaters via digital control from the computer.The temperature inside the chamber was monitored using amercury thermometer. Thermometer values and corre-sponding transducer voltages were recorded with a calibra-tion VI, while the temperature was increased to 100 or to250C (for water and olive oil, respectively) and then de-creased to room temperature. These data were used to esti-mate the duty cycle for the closed-loop controller.

Figure 1. Custom electrospinning system. The electrospinning device consists of the polymer chamber (a), electrodes (b,c),enclosure (d), collection surface (e), and three-axis translation stage (f). [Color figure can be viewed in the online issue, whichis available at www.interscience.wiley.com.]

112 MITCHELL AND SANDERS



4/11

To calibrate the pressure transducer, lab-forced air wasapplied to both the transducer and a vacuum gauge througha medical grade pressure regulator. Pressure was increasedfrom 0 to 172.5 kPa, while voltages were recorded at 34.5 kPaintervals, using a calibration VI on the computer.

Sensitivities for each transducer were determined by us-ing a linear least-squares regression to the voltage verses

measured value data. The sensitivities were used in thecontroller VI for feedback control. The feedback controlswere then evaluated for percent overshoot, time to peak, andsteady state error, under typical operating conditions.

Mesh production and analysis

An analysis of the system parameters was performedusing meshes electrospun from a melt. Melt spinning waschosen over solvent spinning, so as to reduce inconsistenciesrelated to solvent evaporation rates and solution concentra-

tion changes that may result during electrospinning fromsolution.

Controlled parameters

In this electrospinning system, there were three classes ofvariables to consider: the device-specific variables, the poly-mer-specific variables, and the collection surface-specificvariables. The primary device-specific variables were theapplied voltage (V), distance from the electrode to the nozzle(De), nozzle length (NI), and nozzle diameter (Nd) (note: thenozzle is usually referred to as capillary in the electrospin-

ning literature). A consistent NI:Nd ratio of 4.5, consideredoptimal based on previous research,16 was used. The onlypolymer-specific variable considered for melt spinning wasthe polymer viscosity. Viscosity of the melted polymer wasdependent on the melt temperature (T), and since tempera-ture was easily controlled in this system, temperature wasused as the parameter, instead of viscosity. The collectionsurface-specific variables were dielectric strength (DS) andsurface area (SA).

A separate experiment was conducted to determinewhether the viscosity indeed decreased when melt temper-ature increased. Polymer of 5.75 g was placed in a nozzle-shaped chamber of 2 mm diameter and 11.73 mm length.The nozzle was heated in the thermal-controlled melt cham-

ber described earlier and brought up to a specified temper-ature225, 234, or 243Cover a 15-min interval. Polymerwas collected for a 5-min interval and then weighed. Vis-cosity was calculated as the ratio of the shear stress at thenozzle wall to the apparent shear rate at the nozzle wall:

aP

L

R4

2Q

where, P is the pressure gradient across the nozzle (g*(H L)),His the average height of polymer above the nozzle,Lis the nozzle length, R is the nozzle diameter, and Q is thevolumetric flow rate (M/(T)). The polymer density () wasdetermined experimentally by weighing a precise volume ofmolten and condensed polymer, ensuring it was free of all

air bubbles. Three trials were conducted at each tempera-ture. A test statistic of 0.05 was used to make comparisons

between groups.Unique to this device, the distance between the collected

fibers and the nozzle was independent of the electrode po-sition. Thus, the device-specific parameter of collection sur-face distance from the nozzle (Ds) was also considered. Thisparameter was normalized to De for consistent comparisonwith values ofDe (Ds/De ). For analysis of the electro-

spinning parameters of electrode distance (De), ratio of col-lection surface distance to electrode distance (), appliedvoltage (V), and temperature (T), polystyrene Petri dishes(100 mm diameter; 15 mm depth) were used for the collec-tion surface. To determine the effect of the collection surfacedielectric strength (DS), additional tests were conductedusing Teflon, Pyrex, and additional polystyrene Petridishes (0.8 mm thickness and 1.75 mm thickness). More testsusing 150 mm 15 mm polystyrene Petri dishes wereconducted to analyze the effect of collection surface area(SA) on the end product mesh. Table I summarizes theexperimental parameters and the ranges over which theywere varied and analyzed. For most of the parameters, thesevalues also represent the range over which the electrospun

jet was stable enough to collect fibers. The upper and lowerbounds for , V, and Twere determined by systematicallychanging the parameter within a 170-mm electrode spacingrange until a consistent mesh could no longer be collected.The upper and lower bounds for Dewere established usingthe upper and lower bounds for and systematically chang-ing the De until a consistent mesh could no longer be col-lected.

The interactions between variables were assessed. All pa-rameters other than those under study were held constantfor each of these evaluations: The ratio was varied at threelevels within five levels ofDe. Vwas varied at three levelswithin three levels of and three levels ofD

e.Twas varied

at three levels within three levels of . The collection surfacedielectric strength was varied at four levels within threelevels of . The collection surface area was varied at twolevels within three levels of .

The mass flow rate (MFR) was measured for a subset ofthe test parameters. The subsets were as follows: measure-ments were made at five levels ofDe, three levels ofV, threelevels of, and two levels of SA. Thus, a total of 75 inter-actions were evaluated.

Mesh collection

To conduct the analysis, meshes were manufactured fromthermoplastic polyurethane (Estane58315, Noveon, Cleve-

TABLE IAnalyzed Parameter Values

Parameter Analyzed Values

De (mm) 130 150 170 190 210 (mm/mm) 0.39 0.53 0.66V(kV) 25 30 35

T(C) 225 234 243DS (kV) 1 54 84 126SA (mm2) 6082 15,394




5/11

land, OH) while each parameter was varied. For the analysisof the device-specific parameters, to assure that the temper-ature of the melt was consistent throughout testing, thefollowing protocol was used for each test. Approximately5 g of polymer was placed in the electrospinning melt cham-

ber and heated until a slow steady flow emerged from an1-mm nozzle at atmospheric pressure (15 min). Once the

flow was steady, the nozzle tip was cleaned. When a freshdroplet formed on the tip, voltage was applied between thepolymer drop and the positive electrode, electrospinning acontinuous fiber from the nozzle. The continuous fiber wascollected on a surface (inverted Petri dish) that was posi-tioned between the nozzle and the positive electrode. Toanalyze the polymer-specific parameter (related to polymerviscosity), electrospinning was conducted at different poly-mer temperatures.

To ensure a consistent thickness of the sample, duringelectrospinning, the collection surface was translated in thehorizontal plane in concentric circles with diameter decreas-ing at a rate of 5 mm/s. Fiber collection began at the outer

edge of the surface, and then, the circle diameter was de-creased in 2-mm increments until the center of the surfacewas reached. Fiber collection was continued, when the sur-face returned to the original position, in concentric circles ofincreasing diameter. Average sample thickness was 0.3 mm.

To determine the MFR, fibers were collected for 300 s, andthen, the sample mass was measured. Care was taken suchthat all fibers were collected on the surface, throughout thetests. The MFR was the sample mass divided by 300 s.

Mesh analysis

Since the focus of this research was to produce predictableand repeatable scaffolds for biomaterial and tissue engineer-ing applications, an appropriate method needed to be de-veloped to analyze the end product mesh. The properties ofthe mesh of interest were the fiber diameter and inter-fiberspacing. Samples of each manufactured test mesh weremounted on 14-mm OD, 12-mm ID Teflon frames, usingTitebondpolyurethane glue (Franklin International, Co-lumbus, OH). Care was taken to prevent straining the sam-ple during the mounting process. The samples were embed-ded in dyed OCT (Tissue-Tek, Sakura, CA) and flash frozen.The OCT was dyed to enhance the contrast between thetranslucent OCT and the translucent polymer fibers so as tosimplify image processing. For each sample, 20 consecutive,

5-m thick sections were taken perpendicular to the hori-zontal plane of the sample and placed on glass slides forlight microscopy analysis. At least 16 images from the first,last, and two intermediate sections from each sample serieswere taken. The images were digitized using Image ProPlus (Media Cybernetics, Silver Spring, MD) and thenanalyzed using custom algorithms written with the ImageProcessing Toolbox in Matlab (Mathworks, Natick, MA).

To assess fiber diameter, a semi-automated method wasused. The digitized images were imported, converted tograyscale, and thresholded such that only the fiber cross-sections remained in the image. Specs resulting from incon-sistencies in the embedding medium were removed. Todetermine the appropriate threshold level for each image,algorithms in the Matlab image processing toolbox were

used. The algorithms, based on an intensity level specifiedby the user, identified peaks and valleys in the image andfilled in the enclosed regions. Once the user was satisfiedwith the image processing, the properties of each region(fiber) were determined.

The properties of each fiber were determined using theregional properties algorithms in the Matlab image process-

ing toolbox. An ellipse was fit to each fiber cross-section, andthe minor-axis length was determined. The minor-axislength was considered to be the fiber diameter. An averageinter-fiber spacing was determined using the stereologicalrelationship for the mean free distance between particles ina two-dimensional plane, L 0.5(NA)

1/2.40,41 The inter-fiber spacing was defined as the edge-to-edge distance be-tween fibers.

RESULTS

Calibration

Thex-y-zstage has a resolution of 98 m, a repeat-ability of1.3 m, a positional accuracy of 21 m, anda straight-line accuracy of 6.75 m/100 mm on allthree axes. The system had the capability of 60 mm/svelocity with a maximum acceleration of 20 m/s2.Total range of motion was 400 mm on all axes.

The performance of the temperature controller wasdependent on the processing temperature for the poly-mer of interest. Four melt temperatures (180, 225, 234,and 243C) were evaluated in calibration testing. The

time to peak temperature ranged from 180 to 420 s,steady-state error ranged from 0.10 to 0.79C, andpercent overshoot was between 0.10% and 1.03%. Theaverage full-scale output (FSO) error was 0.22% for alltemperatures evaluated.

The pressure controller for a step input from 0.0 to103.0 kPa had a steady-state error of 6.0 kPa, a time topeak of 5 s, and a percent overshoot of 0.01%. Theaverage FSO error for pressure control was 0.41% forall tests conducted.

Parameterization

ANOVAs performed on all data revealed that therewas a significant (p 0.05) effect for all the test pa-rameters with respect to fiber diameter and inter-fiberspacing. In addition, there was no apparent combina-tion of parameters that would reduce the numberneeded for predicting the fiber diameter and inter-fiber spacing features of the end product. Here, pa-rameters are expressed normalized to their maximumtested values. The range of testing within a parameterwas limited by the stability of the electrically spun jet.

For the distance between electrodes (De), fibroporous




6/11

meshes outside of the indicated parameter range (Ta-ble I) could not be collected. Fibroporous meshescould not be collected at voltages (V) less than 25 kV.Voltages above 35 kV resulted in highly unstable jets,which could only be collected at De 170 and 0.39. For all combinations ofDeand V, meshes could

be collected only within the range for indicated inTable I. Temperatures below 225C resulted in poly-mer melts that were too viscous to spin, whereastemperatures above 243C overheated the polymer.The collection surface values were limited by theavailability of acceptable Petri dishes for fiber collec-tion. Data presented here are for the parameter com-bination, within each test variable, that had the great-est influence on the fiber diameter and inter-fiberspacing.

Viscosity testing at controlled temperatures showedthat viscosity decreased with increased temperature.Viscosities at 225, 234, and 243C averaged 678.3(149.9), 136.2 (35.2), and 53.8 (4.5) cP, respec-tively. The viscosity at 225C was significantly higherthan that at 234 or 243C; the viscosity at 234C wassignificantly higher than that at 243C (p 0.05).

Contrary to the data reported in the literature, in-creasing the temperature (T) (which decreased theviscosity) caused an increase in fiber diameter [Fig.2(a,b)]. A concave curve related fiber diameter to ap-plied voltage. Both temperature and applied voltagehad less effect on fiber diameter compared with thefeatures of the collection surface, that is, dielectricstrength and surface area. The dielectric strength (DS)

and surface area (SA) had opposing effects. The diam-eter increased with increasing dielectric strength,whereas the diameter decreased with increasing sur-face area. Fiber diameter was nonlinearly related todistance from the nozzle (De), a parameter previouslyinvestigated for its influence on fiber diameter.10,15

Inter-fiber spacing increased with increasing tem-perature, but was not strongly dependent on appliedvoltage [Fig. 3(a,b)]. The surface area of the collectionsurface had the most significant effect of all parame-ters tested, decreasing with inter-fiber spacing. Thecollection surface dielectric strengths effect was to

increase inter-fiber spacing. Both distance between theelectrodes and the ratio between electrode distanceand collection surface distance exhibited nonlineartrends with respect to inter-fiber spacing.

The complexity of the interdependence of parame-ters is exemplified in Figure 4. Holding the electrodedistance constant and varying the collection surface toelectrode distance ratio resulted in three differenttrends for the dependence of fiber diameter on appliedvoltage. Only the high ratio ( 0.66) showed astrong relationship. The other two ratios ( 0.53 and0.39) had a concave curve and a convex curve, respec-tively.

The average percent change of fiber diameter and

inter-fiber spacing for all parameter tests is shown inFigure 5. Data from all the tests were used to generatethis graph. Contrary to the published literature, ap-plied voltage and viscosity (controlled by tempera-ture) did not have the greatest influence (indicated byaverage percent change) on fiber diameter and inter-

fiber spacing. Within the six parameters of interest, thecollection surface had the greatest impact on fiberdiameter and inter-fiber spacing. The dielectricstrength of the collection surface induced the greatestchange, and the collection surface area inducedslightly less of a change in both parameters.

The effects of the parameters were compared quan-titatively. The average percent change in fiber diame-ter was at least 14 and 15% greater for dielectricstrength and surface area, respectively, than that forany other parameter. For inter-fiber spacing, the in-duced change was at least 34 and 13% greater fordielectric strength and surface area, respectively. To

determine if the difference in percent change was

Figure 2. (a,b) Fiber diameter dependence on manufactur-ing parameters voltage (V), temperature (T), ratio of collec-tion surface distance from nozzle to distance between theelectrodes (), surface area (SA), dielectric strength (DS), anddistance between the electrodes (De).




7/11

significant, Student t tests were performed. The nullhypotheses were that the percent change in fiber di-ameter and inter-fiber spacing induced by one of thefour parameters voltage (V), distance between theelectrodes (De), temperature (T), or the ratio () wasnot different from the change induced by the collec-

tion surface dielectric strength (DS) or surface area(SA). The percent change in fiber diameter induced byDS and SA was significantly (p 0.05) different fromthe change induced byV,T, or . However, it was notsignificantly different from that induced by De. DSinduced a greater change in inter-fiber spacing than V,T, or . The change induced by SA on inter-fiberspacing was not significant when compared with thechange induced by the other parameters.

Though the ANOVA analysis results demonstratedthat there was no combination of parameters thatwould reduce the number needed for predicting thefiber diameter and inter-fiber spacing of the end prod-

uct, it was postulated that MFR was related to at least

some of the test parameters. An increase in MFRwould be expected to increase the fiber diameter.

The MFR decreased slightly with increasing andincreased with increasing voltage. Increasing SA in-creased the MFR, whereas increasing DS resulted in aparabolic relationship with the MFR (Fig. 6).

Correlations between the MFR and fiber diameter,and MFR and inter-fiber spacing were analyzed tofurther investigate the relationships among the pa-rameters. Table II shows the correlation coefficient forMFR verses fiber diameter and inter-fiber spacing for

all test parameters, except one held constant (, DS, orV). MFR was well-correlated to the fiber diameter for

Figure 4. Fiber diameter dependence on voltage for threedifferent values of . Result shows the complexity of theinterdependence of the parameters, with three differentshaped curves depending on the value of .

Figure 5. Relative influence of the six test parameters. Fea-tures of the collection surface, surface area and dielectricstrength, had the greatest influence on fiber diameter andinter-fiber spacing.

Figure 3. (a,b) Inter-fiber spacing dependence on manufac-turing parameters voltage (V), temperature (T), ratio of col-lection surface distance from nozzle to distance between theelectrodes (), surface area (SA), dielectric strength (DS), anddistance between the electrodes (De).




8/11

andVbut not for DS. The inter-fiber spacing corre-lated poorly with MFR for all three test parameters.

Repeatability, predictability, and accuracy

The smallest average fiber diameter achieved for anelectrospun mesh in these initial experiments was5.32 2.71 m, and the smallest average inter-fiberspacing achieved was 4.39 1.41 m. Repeatabilityfor scaffold production had a mean error of 1.8% and20.2% for fiber diameter and inter-fiber spacing, re-spectively. In all repeated sample tests combined,there was no significant difference for diameter (p 0.39) over the range of 8.6314.13 m or for spacing(p 0.20) over the range of 44.7582.85 m.

DISCUSSION

A system that allowed control over a range of elec-

trospinning manufacturing parameters was devel-

oped. A sensitivity analysis was conducted to deter-mine the role of each feature toward fiber diameterand inter-fiber spacing, two features of primary inter-est for biomedical application of electrospun meshes.Such a system has the potential to provide an insightinto why there are conflicting results in the literature

concerning parameter influence on mesh morphology.The system also allowed the repeatability in fiber di-ameter and inter-fiber spacing, under closed-loop con-trol, to be evaluated.

The repeatability for fiber diameter and inter-fiberspacing achieved with this system is far better thanthat reported for other electrospinning systems todate.8,10,17,30,32,33,4244 This result is a reflection of thewell-controlled manufacturing system used. Further,the result here, all test parameters having a significanteffect on fiber diameter and inter-fiber spacing, pointsto the importance of system control in this application.A lack of thorough system control may help to explaindiscrepancies in the literature concerning how manu-facturing parameters affect scaffold architecture.

A total of 75 interactions among parameters wereinvestigated here, and thus, the parameter space waswell-studied. However, to more thoroughly character-ize the dependence of specific parameter sets on fiberdiameter and inter-fiber spacing and to thus generatemore data points for Figures 24, and 6, additionalstudies would need to be carried out.

Fiber diameter range and inter-fiber spacing rangedepend on the polymer selected, the viscosity controlmethod (melting) used, and the collection surface ma-

terial. Different selections, for example Poly(l-lactide,-caprolactone) with a solvent, allow for smaller fiberdiameter ranges,45 more consistent with those re-ported to reduce fibrous encapsulation in vivo.3

Contrary to the literature,10,25 in this investigation,the fiber diameter increased when the applied voltagewas increased. This discrepancy is an expected reflec-tion of the closed system used here, as opposed to theopen systems described in previous research. The typeof system affects the relationship between appliedvoltage and MFR. It is a general consensus amongresearchers that increasing the applied voltage in-

creases the force on the polymer, which in turn de-forms the droplet of polymer and initiates fiber spin-ning. It has also been reported that increasing the

TABLE IICorrelations of Fiber Diameter and Inter-Fiber Spacing

with Mass Flow Rate with All ParametersExcept One Held Constant

Parameter not HeldConstant Fiber Diameter

Inter-FiberSpacing

0.97 0.39DS 0.21 0.06

V 0.84 0.35

Figure 6. Mass flow rate dependence on voltage (V), ratioof collection surface distance from nozzle to distance be-tween the electrodes (), surface area (SA), dielectricstrength (DS), and distance between the electrodes (De).




9/11

voltage/force increases the MFR of the polymer16,17

and also increases the strain rate on the polymer.16

Thus, if the flow is limited, as in a closed system, thenthe polymer strain rate increases with applied voltage,and smaller fibers will result. However, in an opensystem, if the voltage is increased, the resulting fiber

diameter will depend on the viscosity of the solution.With high-viscosity solutions, the frictional forces willlimit the flow, resulting in an increased strain rate andsmaller fiber diameters with increased voltage. But,with low-viscosity solutions, as used here, the flowrate increases while the strain rate remains constant,resulting in larger fiber diameters with increased volt-age.

Results here for the correlation between polymertemperature (inversely related to viscosity as con-firmed by the viscosity tests reported here) and fiberdiameter are also inconsistent with the litera-ture.35,39,46,47 It is reasonable to expect the fiber diam-eter to decrease with increased temperature, becausethe polymer viscosity is decreased, and thus, the strainrate is increased. However, if the viscosity becomessufficiently low, the frictional forces in the fluid are nolonger a factor, and the MFR will increase with in-creased temperature, resulting in an increase in fiberdiameter. This explanation is confirmed by the highcorrelation shown in Table II between MFR, fiber di-ameter, voltage, and relative distance from the nozzle.

Consistent with reports in the literature,10,15 fiberdiameter increased with increasing distance from thenozzle. However, unlike in the literature, the decay

was not asymptotic. The difference here is expected, aresult of where the analysis was conducted. Previousreports in the literature have focused on analysis ofthe stable region of the jet, not the end product mesh.Fiber diameter is reported to reduce in excess of 100times9 from the jet secondary bending stability to thecollection surface, in effect, removing the asymptoticdecay characteristic. Thus, the lack of an asymptoticrelationship here is not necessary in conflict with theliterature.

It is reasoned that the collection surface had themost significant effect on fiber diameter of the param-

eters investigated because of its effect on the electricfield. The system described here was unique in thatthe collection surface was independent of the elec-trodes, and the collection surface was positioned be-tween the electrodes. The collection surface likely dis-torted the electric field (Fig. 7), causing it to deflectaround the surface, and thus, causing a change involtage. The interdependence of the parameters col-lection surface dielectric strength, distance betweenthe electrodes, and applied voltage on fiber diametersupports this interpretation. The electric field distor-tion is most likely responsible for the collection surfacedominance in the sensitivity analysis. If the electro-

spinning device is viewed as a closed-loop system,

then the current must flow from the ground electrodethrough the jet and collection surface to the positiveelectrode. Increasing the dielectric strength in this sys-tem configuration increased the resistance betweenthe electrodes, decreasing the current drawn betweenthe electrodes. Thus, the work available to strain thepolymer was reduced. The result was an increase infiber diameter.

The physical size of the collection surface also af-fected the electric field between the electrodes. Placinga dielectric between the electrodes deflected the elec-tric field. Increasing the surface area of the collectionsurface increased the degree of field deflection. Thiswould tend to increase the secondary instability of thespinning jet, which would result in smaller fiber di-ameters. In addition, increasing the secondary insta-bility (hence wave propagation frequency) decreasedthe mesh inter-fiber spacing, as confirmed in Figure 2.

The repeatability achieved with this system, 1.8%for fiber diameter and 20.2% for inter-fiber spacing, isexpected within the needs for tissue engineering ap-plications. Consistent fiber diameters may be impor-

tant to tissue response,3 and a consistent architecturewill ensure meshes with relatively uniform mechani-cal properties without locally weak regions. A nextstep will be to conduct implant testing to evaluate invivo performance using the meshes as coatings forbiomaterial devices, scaffolds for tissue engineeringapplications, or transporters for large populations ofcells.

CONCLUSIONS

A controlled electrospinning system was fabricated

to evaluate the sensitivity of fibro-porous mesh archi-

Figure 7. Expected field lines between electrodes. The field

lines distort in the presence of a collection surface (rightimage) compared with that when no collection surface ispresent (left image).




10/11


11/11

32. Deitzel JM, Kleinmeyer J, Harris D, Tan NCB. The effect of

processing variables on the morphology of electrospun nano-

fibers and textiles. Polymer 2001;42:261272.

33. Zong X, Kim K, Fang D, Ran S, Hsiao BS, Chu B. Structure and

process relationship of electrospun bioabsorbable nanofiber

membranes. Polymer 2002;43:44034412.

34. Lu Q-W, Hernandez-Hernandez ME, Macosko CW. Explaining

the abnormally high flow activation energy of thermoplasticpolyurethanes. Polymer 2003;44:33093318.

35. Demir MM, Yilgor I, Yilgor E, Erman B. Electrospinning of

polyurethane fibers. Polymer 2002;43:33033309.

36. Fridrikh SV, Yu JH, Brenner MP, Rutledge GC. Controlling the

fiber diameter during electrospinning. Phys Rev Lett 2003;90:

14 (144502).

37. Hohman MM, Shin M, Rutledge G, Brenner MP. Electropin-

ning and electrically forced jets. II. Applications. Phys Fluids

2001;13:22212236.

38. Larrondo L, St John Manley R. Electrostatic fiber spinning from

polymer melts. III. Electrostatic deformation of a pendant drop

of polymer melt. J Polym Sci Polym Phys Ed 1981;19:933940.

39. Liu H, Hsieh Y-L. Ultrafine fibrous cellulose membranes from

electrospinning of cellulose acetate. J Polym Sci Part B: Polym

Phys 2002;40:21192129.

40. Russ JC, Dehoff RT. Practical Stereology, 2nd ed. New York:Kluwer Academic; 2000. p 127.

41. Underwood EE. Quantitative Stereology. Reading: Addison-Wesley; 1970. p 148188.

42. Lee KH, Kim HY, Khil MS, Ra YM, Lee DR. Characterization ofnano-structured poly(epsilon-caprolactone) nonwoven matsvia electrospinning. Polymer 2003;44:12871294.

43. Buer A, Ugbolue SC, Warner SB. Electrospinning and proper-ties of some nanofibers. Text Res J 2001;71:323328.

44. Theron SA, Zussman E, Yarin AL. Experimental investigationof the governing parameters in the electrospinning of polymersolutions. Polymer 2004;45:20172030.

45. Mitchell SB. Electrospinning controlled architecture scaffoldsfor tissue engineering and the effect of scaffold mechanicalproperties on collagen synthesis in tissue engineered mitralvalves. Doctoral Dissertation, University of Washington, 2004.

46. Mo XM, Xu CY, Kotaki M, Ramakrishna S. Electrospun P(LLA-CL) nanofiber: A biomimetic extracellular matrix for smoothmuscle cell and endothelial cell proliferation. Biomaterials2004;25:18831890.

47. Boland ED, Wnek GE, Simpson DG, Pawlowski KJ, Bowlin GL.Tailoring tissue engineering scaffolds using electrostatic pro-

cessing techniques: A study of poly(glycolic acid) electrospin-ning. J Macromol Sci Pure Appl Chem 2001;38A:12311243.


dielectric mijloc intreelectrozi

Documents