Abstract Experimental measurements are presented to provide phe- nomenological insight into the commercial melt blowing process. In particular, we discuss the following experimental measurements obtained at various die-collector locations: fiber diameter, fiber velocity, air velocity, fiber acceleration, fiber entanglement, fiber temperature, birefringence, wide- angle x-ray diffraction and small-angle x-ray scattering. Our discussion focuses on how these meas
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Abstract Experimental measurements are presented to provide phe-nomenological insight into the commercial melt blowingprocess. In particular, we discuss the following experimentalmeasurements obtained at various die-collector locations:fiber diameter, fiber velocity, air velocity, fiber acceleration,fiber entanglement, fiber temperature, birefringence, wide-angle x-ray diffraction and small-angle x-ray scattering. Ourdiscussion focuses on how these measurements provideinsight into fiber formation during melt blowing. Introduction Figure 1 illustrates the basic melt blowing (MB) process. Athermoplastic fiber-forming polymer is extruded throughsmall orifices into convergent streams of hot air that rapidlyattenuate the extrudate into small diameter fibers. The airstreams also transport fibers to a collector where they bond atfiber-fiber contact points to produce a cohesive nonwovenweb. Even though MB technology wasdeveloped nearly 50 years ago [1], anunderstanding of the MB process has been elusive because fiber motionduring MB is extremely complex.However, studies of laboratory scalesingle-hole MB during the pastdecade [2-10] have revealed impor-tant information about fiber flow inhigh-velocity air streams. The commercial MB process is morecomplex than single-hole MB sincethe commercial process includesnumerous die orifices that produce acomplex fiber stream with extensive fiber entanglement.Studies of multi-orifice MB operating at commercial speed[11-13] have shown that fiber entanglement (i.e. developmentof web structure) begins as close as 1 cm to the die. However,web structure is dynamic and becomes fixed only when fiberscontact the collector. Multi-orifice studies have shown thatfiber entanglement affects fiber flow (e.g. fiber speed unifor-mity, fiber diameter attenuation and fiber orientation) as wellas web properties (e.g. filtration performance, basis weightuniformity and web stiffness). Consequently, fiber flow in thewhole die-collector region and the phenomenon of fiberentanglement must be considered to understand web struc-ture development during MB. Several years ago, we began an experimental campaign toobtain basic information about the commercial multihole MBprocess. We have reported information about air speed, airtemperature, fiber speed, fiber temperature, fiber acceleration,fiber diameter attenuation, fiber entanglement and fiber lay- Fiber Formation DuringMelt Blowing By Randall R. Bresee and Wen-Chien Ko, Department of Materials Science and Engineering,University of Tennessee, Knoxville, TN 37996-2200 ORIGINALPAPER/ PEER-REVIEWED 21 INJ Summer 2003 Figure1 ILLUSTRATION OF THE BASIC MB PROCESS  down on the collector. In this paper, we will discuss many of these measurements to summarize basic phenomenologicalevents occurring during commercial MB. In particular, we willemphasize the development of fiber structure during MB anddiscuss basic differences between MB and conventional fibermelt spinning (MS). Experimental Procedures We processed a Ziegler-Natta catalyzed polypropylene (PP)resin (1259 MFR) provided by ExxonMobil ChemicalCompany. Melt blowing was performed with an AccurateProducts 600-hole (51 cm wide) horizontal line in TANDEC atthe University of Tennessee using processing conditions simi-lar to conditions used for commercial web manufacture. Thedie hole diameter was 0.0145 inch, die set back/air gap dis-tances were 0.060 inch, primary air pressure at the die was 4.0psi and the mass throughput rate of the resin was 0.4 ghm.The die-to-collector distance (DCD) was 30 cm unless the col-lector was moved as far as possible from the die (90 cm) to rep-resent the absence of a collector. Adigital infrared thermometer with laser sighting andadjustable emissivity (Cole Parmer Model 39650-20) was usedto measure fiber temperature on-line. Thermal emissivity wasset at 0.95 and measurements were obtained from a circulararea of 2.0 cm diameter. Fibers were concentrated in this area by inserting a probe of 2.5 cm diameter into the fiber streamand obtaining measurements from the fibers as they accumu-lated on the surface of the probe. Temperature values report-ed here were the maximum values observed during a 10 s timeinterval. We thought this procedure provided good estimatesof fiber temperature since the probe was a poor heat conduc-tor, fibers from the multihole MB line accumulated on theprobe quickly and the response time of the thermometer wasreasonably fast (500 ms). This assumption was supported bythe observation that temperature readings typically attainedvalues close to their maximum value within 1-2 s of fiber accu-mulation and then remained fairly constant thereafter. Air temperature and air velocity in the machine direction(MD) were measured at various locations on-line using a com-mercial multifunction handheld instrument containing a plat-inum resistor for temperature and a Pitot-static tube for veloc-ity (maximum measurable velocity was 135 m/s). These mea-surements were obtained using the same processing condi-tions as those used for fiber measurements but with no resinthroughput and readings were obtained at centerline posi-tions where air temperature and air velocity were maximum.The temperature and velocity values reported here were max-imum values observed during a 60 s time interval. Arapid framing rate camera and pulsed laser were used toacquire high-speed images on-line. The hardware consisted of a Phantom Arrow digital camera (512 x 512 x 8-bit, 1000frames/s) and Oxford Lasers HSI1000 Illumination Systempurchased from Vision Research. Fiber velocity and fiberacceleration were obtained by analysis of triply exposed indi-vidual images using known time intervals between laser puls-es and known spatial dimensions of images [11]. Diameters of single fibers were measured two ways.Diameters larger than approximately 60 µ m were measuredfrom individual high-speed images acquired during MB. Fiberdiameter measurements smaller than 60 µ m were obtained off-line using WebPro, an automated image analysis system [14]from fibers retrieved at various locations between the die andcollector.Fiber birefringence was measured off-line using polarizedoptical microscopy and a single-order compensator. Wide-angle x-ray diffraction (WAXD) measurements wereobtained using a rotating anode x-ray source and a two-dimensional detector. Small-angle x-ray scattering (SAXS) wasmeasured using a 10 m instrument at Oak Ridge NationalLaboratory [15]. Results and Discussion In this paper, we will focus on three important aspects of theMB process in an attempt to understand fiber formation dur-ing MB. These are fiber velocity and acceleration, fiber diame-ter attenuation, and development of fiber morphology (poly-mer crystallization and molecular orientation). Fiber Velocity and Acceleration Figure 2 provides on-line measurements acquired in themachine direction (MD) during MB when a collector wasabsent. This figure shows that mean fiber speed was very slowat the die but fibers quickly accelerated to higher speed. Meanfiber acceleration beyond about 2 cm from the die decreasednearly as rapidly as it had increased so maximum fiber speedwas attained about 6 cm from the die. Figure 2 indicates thatthe MB process is not particularly fast in terms of maximumfiber speed when compared to fiber MS (60m/s during MB iscomparable to fiber speed during MS). Since MB collectors aretypically placed 25-30 cm from the die, this figure shows thatmean fiber velocity increased through only a relatively smallportion of the die-collector space and decreased through mostof the die-collector space, although it remained relatively large 22 INJ Summer 2003 Figure2 ON-LINE MEASUREMENTS OF MEAN FIBERVELOCITYIN THE MD, MEAN AIR VELOCITYIN THE MD AND MEAN FIBER ACCELERA-TION IN THE MD WHEN ACOLLECTOR WASABSENT  at locations where a collector is typically placed. It is important to recognize that fiber velocity and accelera-tion profiles for MB are fundamentally different than the S-shaped velocity profile and hat-shaped acceleration profileobserved for MS. For example, the steady-state mass balancecommonly used to relate fiber diameter to fiber speed duringMS (continuity equation) is largely inapplicable to MB because steady-state conditions are not achieved during MBexcept possibly close to the die (fiber speed is substantiallytime dependent at most die-collector locations). Consequently,it is inappropriate to estimate MB fiber speed from fiber diam-eter measurements except near the die. This is obvious in lightof the observation that fiber diameter attenuation occurs at alldie-collector locations even though average speed decreases atmost die-collector locations. Figure 3 provides high-speed images acquired to visualizedynamic aspects of the MB process when a collector wasabsent [11]. Images in this figure were reproduced from high-speed digital cines acquired near the die, and 4, 9 and 19 cmfrom the die. The MD is left-to-right in each image and the airknife gap can be seen near the left edge of the image acquirednear the die.These images provide much information about the physicalstatus of fibers during MB and show that MB is fundamental-ly different than MS. Unlike MS for example, the amount of fiber length per unit volume of space increases substantiallyduring MB with increasing distance from the die. Also unlikeMS, fiber orientation changes substantially during MB fromstrongly MD near the die to isotropic 9 cm from the die to sub-stantially cross-machine direction (CD) 19 cm from the die.Finally, Figure 3 shows that fiber entanglement increases withincreasing distance from the die. Overall we see that differ-ences in fiber length per unit volume of space, fiber orienta-tion and fiber entanglement make the commercial MB processfundamentally different and substantially more complex thanthe MS process. We believe that the basic cause of these differ- 23 INJ Summer 2003 Figure3 FIBERS NEAR THE DIE, AND 4, 9 AND 19 CM FROM THE DIE WHEN ACOLLECTOR WASABSENT [11]; IMAGE AREA= 19 MM X 19 MM NEAR THE DIE AND 17 MM X 17 MM FOROTHER LOCATIONS  ences is that fiber velocity decreases through much of the die-collector space during MB whereas fiber velocity does notdecrease during MS. Fiber Diameter Attenuation One of the distinctive features of the MB process is the easewith which fine fiber diameters may be produced. Figure 4 provides mean fiber diameter measurements at various loca-tions between the die and collector. This figure shows thatfiber diameter attenuation was extremely rapid near the diesince fiber diameter was reduced to about 15% its srcinal size(from 368 :m to about 60 µ m) after traveling less than 1 cmfrom the die. Figure 2 showed that air velocity was maximumand fiber velocity was minimum at the die so it is likely thataerodynamic drag near the die provided the elongationalforce necessary to attenuate fiber diameter in this region. Figure 2 also showed that mean air and fiber velocities in theMD became nearly equal at about 10 cm from the die so weshould expect conventional drag force in the MD to be rela-tively small beyond 10 cm from the die. However, fiber diam-eter attenuation continued to occur far from the die eventhough mean drag force in the MD is not expected to be large.This is shown in Figure 5 where measurements from the pre-vious figure are replotted so fiber diameter attenuation farfrom the die is more easily observed. This figure implies thatan elongational force other than conventional aerodynamicdrag near the die may exist far from the die. What phenome-na could possibly provide such an elongational force? Figure 6 shows maximum, minimum and mean fiber accel-eration values among twenty measurements obtained at vari-ous die-collector locations. This figure shows that the MBprocess is characterized by a distribution of fiber accelerationvalues that is remarkably broad at most locations between thedie and collector. For example, acceleration values for indi-vidual fibers at 9 cm from the die ranged from -122,000 to+57,000 m/s/s. Figure 6 also shows that positive accelerationvalues were observed at all die-collector locations eventhough average fiber acceleration was negative at many die-collector locations. This suggests that fibers may collide like“bumper cars” during MB to accelerate and decelerate indi-vidual fiber segments [11]. It is reasonable to expect that afiber segment accelerated in this manner could elongate if itwere constrained by entanglement with other fibers. Sinceentanglement begins within 1 cm of the die, it is possible thatthe phenomenon of fiber contact/entanglement elongatesfibers and contributes to diameter attenuation at most die-col-lector locations. Figure 3 showed that fiber orientation far from the die waspredominantly CD even though it is well known that MBwebs usually exhibit preferred MD orientation,. Since Figure 3 was obtained when a collector was absent but typical MBwebs are produced with a collector, we must conclude that the 24 INJ Summer 2003 Figure4 MEAN FIBER DIAMETER DURING MB WHEN ACOLLECTOR WAS ABSENT Figure5 MEAN FIBER DIAMETER FAR FROM THEDIE WHEN ACOLLECTOR WAS ABSENT Figure6 MAXIMUM, MEAN AND MINIMUM FIBERACCELERATION MEASUREMENTS IN THEMD WHEN ACOLLECTOR WAS ABSENT Distance from Die (cm)Distance from Die (cm)    F   i   b  e  r   D   i  a  m   e   t  e  r   (      µ   m   )      A  c  c  e   l  e  r  a   t   i  o  n   (   1   0   0   0  m    /  s    /  s   )
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks