AbstractPolybutylene (PB-1) fibers were spun at spinning speeds of

250-2500 m/min. A tensile tester was used to analyze thestress-strain behavior of these fibers. In addition, birefrin-gence and the effect of aging were examined. A DMA(dynamic mechanical analyzer) was used to measure the stor-age modulus and loss modulus of the fibers. Nonwoven matsof the fibers were prepared and compression tests were run onthese mats. The properties of the polybutylene fibers andmats were compared with the properties of commonpolypropylene fibers and mats.

IntroductionFor reasons of relative simplicity and high speed, melt spin-

ning is the most widely used industrial process for fiber for-mation. In melt spinning, polymer is extruded through a cap-illary (spinneret) in order to form fibers. Mechanical rolls areused as the external force that draws and attenuates themolten polymer into fibers. The polymer molecules and/orcrystals orient parallel to the fiber axis and thus impart spe-cific characteristics (e.g., strength) to the final filaments. Thefibers are generally wound up on bobbins for use in furtherprocessing (e.g., weaving, knitting, composite layup, etc.). Aprocess called spunbonding, which is also of great industrialimportance, is quite similar to melt spinning except that an airventuri, rather than a mechanical roll, provides the attenuat-ing force. In spunbonding, the fibers are collected as a non-woven web below the venturi. Typical commercial melt spin-ning and spunbonding speeds are in the range of 2000 to 6000m/min.

Poly(1-butene)Polypropylene is a versatile, inexpensive polymer that is

widely used in fiber production as well as being used formolding, sheeting, and a myriad of other products. A muchless common polyolefin is poly(1-butene), which is oftenreferred to as PB-1, polybutene-1, or simply polybutylene.Polybutylene fiber spinning is the subject of this paper.Polybutylene has properties that are typical of other poly-olefins, such as a good resistance to chemical attack.However, polybutylene has some characteristics that differfrom other polyolefins. For example, polybutylene has lowcold flow and high hoop strength. Hence, polybutylene issuitable for use in piping; see Schemm (2002).

Natta and Corradini (1956) and Natta et al. (1960) first stud-ied polybutylene (PB-1) and its crystallization process. Theyshowed in their studies that PB-1 exists as two different poly-morphs. The first polymorph is an unstable structure thatthey called “Form II”; Form II consists of a tetragonal struc-ture. Natta’s group found that Form II slowly transforms intoa more stable Form I, which is a hexagonal structure with unitcells containing a 3-fold helical conformation that resemblesthe cell structure of polypropylene. A third polymorph (FormIII) was noted by Zannetti et al. in 1961; this polymorph formsfrom solution. In 1964, Holland and Miller discussed“untwined” hexagonal crystals (Form I’) that can also formfrom solution. The known polymorphs of PB-1 are summa-rized in Table 1.

Boor and Mitchell (1963) and Choi et al. (1966) carefullyexamined the time required for transition from Form II toForm I. Boor and Mitchell gave the transition halftime (atroom temperature) as about 250 minutes. In a 28 day test,Choi et al. observed the density of a PB-1 sample rise from0.895 to 0.910 and the crystallinity rise from 25 to 55%. Weretaand Gogos (1971) analyzed the crystal transition while thesample was under deformation.

Abenoza and Armengaud (1981) used laser Raman spec-troscopy to examine the crystal forms of PB-1. Weynant et al.(1982) examined PB-1 crystals subjected to uniaxial tension,while Nakafuku and Miyaki (1983) examined the effect ofpressure on PB-1 crystallization.

Melt-Spun Polybutylene Fibersand NonwovensBy Diana L. Ortiz and Robert L. Shambaugh*, School of Chemical Engineering and Materials ScienceUniversity of Oklahoma, Norman, Oklahoma

ORIGINAL PAPER/PEER-REVIEWED

36 INJ Winter 2005

* Correspondence: Telephone: (405) 325-6070 Fax: (405) 325-5813 email: shambaugh@ou.edu

Hong and Spruiell (1985) processed PB-1 by cold rolling andfilm blowing. They then used density and wide angle X-raytechniques to show that applied stresses enhance the Form IIto Form I transition. Shaw and Gilbert (1991) produced com-pression-molded samples from 8 different grades of materialcontaining PB-1 and other olefins; in a later paper (1994) theyalso produced fibers from these 8 polymers. They showedthat the presence of other olefins (as a comonomer or a blend)affects the sample crystallinity and, hence, the physical prop-erties of the sample. Choi and White (1998) did furtherdetailed work on the structure development during melt spin-ning of PB-1.

Nakamura et al. (1999) fabricated films from PB-1 andobserved that stress causes Form I to appear immediately dur-ing processing. Samon et al. (2000; 2001) examined the struc-ture of PB-1 fibers as they were being produced and after(room temperature) annealing; they used spinning speeds upto 250 m/min. They found that higher spinning speedscaused an increase in crystallinity during the annealing step.

Our study analyzed the effect of different processing condi-tions on the final properties of polybutylene fibers producedat spinning speeds up to 2500 m/min -- an order of magnitudehigher than has been previously examined. The properties ofthe aged versus freshly produced fibers were compared. Also,comparisons of solid versus hollow PB-1 fibers and fiber matswere done. Hollow fibers were included in the study becauseprevious work (de Rovere and Shambaugh, 2001) with PPshowed that the properties of hollow fibers were comparableto the properties of solid fibers.

Experimental DetailsGrade 0400 polybutylene-1 (Basell Polyolefins) was used in

our work. This PB-1 has a melting temperature of 123.9 –126.1 OC, a specific gravity of 0.915, and a melt flow rate (MFR)of 20. The melt spinning of the polybutylene was carried outusing two types of spinning equipment. The first type (seeFigure 1) consisted of a Brabender™ single screw extruder, aspin pack equipped with a gear pump, and a spinneret. The

extruder barrel was 19.0 mm (0.75 in.) in diameter and 381 mm(15 in) in length and had a compression ratio of 3:1. The spinpack contained a Zenith™ gear pump that metered and pres-surized the molten polymer. A single-hole spinneret was usedfor the production of solid fibers, and a tube-in-orifice spin-neret was used for producing hollow fibers. The single-holespinneret had a diameter of 0.407 mm (0.016 in) and a length2.97 mm (0.1169 in). The tube-in-orifice spinneret had a poly-mer annulus with a 1.98 mm (0.078 in) outer diameter and aninner diameter of 1.22 mm (0.048 in). The nitrogen capillarydiameter was 0.76 mm (0.030 in).

The polymer filament was drawn by a mechanical take uproll located 1 m below the spinneret. This roll was used onlyfor spinning speeds up to 1500 m/min. For higher spinningspeeds an air venturi was used. This venturi allowed spin-ning speeds up to 4000 m/min. The Brabender extruder waskept at 225 OC, the spin pack was kept at 205 oC, and the spin-neret was kept at 190 OC.

The continuous extruder system shown in Figure 1 was

37 INJ Winter 2005

Figure 1THE EXPERIMENTAL EQUIPMENT USED TO

SPIN FIBERS

Table 1THE DIFFERENT POLYMORPHS OF POLYBUTYLENE.

The data are from Choi et al. (1966) and Nakamura et al. (1999).

used for the higher polymer flow rates (0.45, 0.50, and 1.0g/min). For lower polymer flowrates (0.23 and 0.30 g/min),a ram extruder was used to provide smooth, controllablepolymer flow. This equipment operated in batch, rather thancontinuous mode. The ram extruder barrel had a length of 8in (203 mm) and a 0.375 in (9.525 mm) diameter. The samespinnerets that were used for the continuous system (Figure 1)were also used for the batch system. For the ram extruder, theextruder barrel was kept at 225 OC and the spinneret was keptat 190 OC.

Offline diameter measurements on the polybutylene fiberswere done with a Nikon Labphoto2 POL Optical Microscopeequipped with a micrometer eyepiece. Cross-sectional areasof hollow fibers were determined by microtoming the fibersand then examining them under the microscope.Birefringence measurements were also done with this micro-scope. Birefringence is defined as the difference between theparallel and the perpendicular refractive indexes of the fiber.Specifically,

(1)

where Δ is the birefringence and nparallel and nperpendicularare, respectively, the refractive index in the parallel and per-pendicular directions to the fiber axis. With round fibers, fiberbirefringence can be determined by the equation

(2)where Δ is the relative retardation in nm, and d is the fiber

diameter in nm. See Phillips (1971) for more information onthese equations. In our work, the relative retardation wasmeasured using the polarizing microscope with a Sernamontcompensator technique.

The mechanical properties of the fibers were measured withan Instron model TT-B-L tensile tester. Each fiber sample wascarefully placed between pneumatic grips; a gauge length of2.3 cm was used. The fiber was then stretched at a constantcross-head speed of 2.54 cm/min (1.0 in/min) until breakageoccurred. The stress-strain curve allowed the determination ofultimate yield strength, elongation at break, and modulus ofelasticity.

Dynamic mechanical analysis (DMA) was used to obtainlinear viscoelastic properties of the fibers as a function of tem-perature. The equipment used for this analysis was aRheometric Scientific model RSA II. Individual fiber sampleswere held by monofilament grips and subjected to a preloadof 1 gram. The storage and loss moduli were measured at afixed frequency of 10 rad/sec (1.59 Hz) within a temperaturerange of -100 °C to 120 °C. Increments of 1.5 °C were usedfrom -40 to 0 °C, and 4 °C increments were used elsewhere inthe temperature range.

Compression tests were run on nonwoven webs of ourpolybutylene fibers. These webs were produced by collectingthe fibers on a metal screen that was placed 40 cm below theair venturi. The compression tests were performed with amodification of standard test IST 120.4 (95) from the

Association of the Nonwoven Fabrics Industry (StandardMethods, 1995). This test uses weights and plates to deter-mine the compression properties of the nonwoven material.Figure 2 shows a schematic of the set up. This modification ofthe standard test was originally developed by de Rovere andShambaugh (2001) for use on webs of polypropylene fibers. Inthis modification, the sizes of the nonwoven specimens arescaled to 25% of the original values given in the INDA stan-dard; however, the applied pressure (1825 Pa) on the speci-mens is identical to that suggested in the standard.

To provide comparisons with polypropylene spinning, weused 88 MFR Dypro isotactic polypropylene donated by theFina Company. This polypropylene has an Mw of 165,000 anda Mn of 41,500. The same equipment used to spin the poly-butylene was used to spin the polypropylene, and spinningconditions (except for operating temperatures) were the sameas for the polybutylene.

Results and DiscussionProperties of Aged and Fresh Polybutylene Fibers

Figure 3 shows results of tensile tests run on fresh PB-1fibers spun using a polymer flow rate of 0.5 g/min and spin-ning speeds ranging from 500 to 2500 m/min. By “fresh” ismeant fibers whose tenacity was determined within 4 hoursafter spinning. The final fiber diameters ranged from 37.5microns for the 500 m/min speed to 16.7 microns for the 2500m/min speed. The highest take-up velocity gave the highestultimate tensile strength and the lowest elongation; suchbehavior is common in polymer spinning (Bansal andShambaugh, 1998a, 1998b; Vassilatos et al., 1985).

Some of the fresh fibers were set aside and aged; Figure 4shows the stress-strain behavior for these aged PB-1 fibers.“Aged” is defined as fibers that are allowed to sit at room tem-perature for one week. Like the fresh fibers, the aged fibersexhibit low tenacity and high elongation at the 500 m/minspinning speeds, intermediate stress-strain behavior at the1000 m/min speed, and high tenacity with low elongation atspeeds of 1500-2500 m/min. Both fresh and aged fibers showslight increases in tenacity and decreases in elongation in therange 1500-2500 m/min. The breaking strength (tenacity) ofthe fibers is of use in quantifying the differences between the

38 INJ Winter 2005

Figure 2THE APPARATUS USED FOR

COMPRESSION TESTS

larperpendicuparallel nnn −=Δ

dR

n =Δ

aged and fresh fibers; Table 2 summarizes the tenacities offresh and aged fibers. The aged samples may appear slightlyweaker than the fresh fibers for spinning speeds of 500, 1000,1500, and 2100 m/min; however, statistical calculationsshowed that, within a 95% confidence level, the tenacity mea-surements of the fresh versus aged samples are not signifi-cantly different. The standard deviations for these measure-ments are given in Table 2.

Figure 5 shows the final diameter (the diameter at the col-lector) of polybutylene fibers as a function of spinning speed.Data are given for polymer flowrates from 0.23 to 1.00 g/min.Data for aged and fresh samples are also shown in Figure 5.As is typical for fiber spinning, smaller fiber diameters resultfrom increased spinning speed and/or decreased polymer

throughput. The diameters can also be predicted with thecontinuity equation that for this spinning system is

(3)

whereρ is the fiber density in g/m3

m is the polymer mass flow rate in g/mind is the fiber diameter in mv is the fiber velocity in m/minThe continuity predictions of the fiber diameters corre-

spond well with the experimental data. Figure 5 also showsthat the aged and fresh samples have almost the same diame-

39 INJ Winter 2005

Table 2TENACITY AND STANDARD DEVIATION FOR FRESH AND AGED PB-1 FIBERS SPUN AT 0.5

G/MIN AND DIFFERENT SPINNING SPEEDS.

)4

(v

md

⋅⋅=

πρ

Figure 3TENSILE TESTS RUN ON FRESH POLYBUTYLENE FIBERS SPUN USING A FLOWRATE OF 0.5

G/MIN AND SPINNING SPEEDS FROM 500 TO 2500 M/MIN.Final fiber diameters varied from 37.6 microns for a spinning speed of 500 m/min to 16.8 microns for a spinning speed of2500 m/min. Each curve is the average of 5 to 8 individual tensile tests. Of the 5-8 fibers involved in these tests, about halfwere produced during one day’s run, while the other half were produced on another day. There was essentially no differ-

ence between the tensile results for the fibers made on different days.

ter profile. Since the standard deviation of the diameter mea-surements is about 1.2 microns, there is no statistical differ-ence between the diameters of the aged and fresh fibers. So,for melt-spinning of polybutylene fibers, the diameter of thefilaments does not significantly vary during the agingprocess.

Figure 6a shows the breaking strength (tenacity) of agedpolybutylene fibers for polymer flow rates of 0.23 to 1.00grams/min. As expected, the tenacity of the fibers increasesas the spinning speed is increased; this behavior is approxi-mately linear. Also, lower polymer flow rates give higherfiber tenacities at the same spinning speed. This effect is like-ly due to higher molecular orientation induced by the higherdraw ratio. The tenacity values in Figure 6a are quite high:fully drawn polyester and polypropylene have tenacities inthe same range as the upper curve in the figure.

To compare the tenacities of aged fibers with the tenacitiesof fresh fibers, the best-fit lines from Figure 6a were replottedon Figure 6b. Also on Figure 6b are data for the tenacities offresh fibers produced at polymer flow rates of 0.5 and 1.0g/min. These fresh fiber samples were tested within 4 hoursafter the fibers were melt spun. The data for aged samples inFigure 6a were produced by setting aside fibers from the same

batches of fibers that were used as fresh samples in Figure 6b.As stated earlier, the aged samples were allowed to age atroom temperature for one week before the tenacity tests wereconducted. Figure 6b shows that the fresh and aged sampleshave very similar tenacities. A quantification of this similari-ty is given in Table 2. This table compares the tenacities andstandard deviations of the tenacities for aged and fresh sam-ples produced at 0.5 g/min. Statistical analysis (t-test) showedthat there is essentially no difference in the tenacities of theaged versus the fresh samples. The same results apply for apolymer rate of 1.0 g/min.

Fiber ModulusBesides the tenacity, the fiber modulus was also deter-

mined. For the same aged fibers that are the subject of Figure6a, Figure 7a shows the moduli plotted as a function of spin-ning speed. The modulus was calculated as the slope of theinitial linear segment in the stress-strain curve (specifically,the modulus was calculated as the slope at a strain of zeropercent).

Figure 7a shows that, for each of the flow rates, the modulusincreases with an increase in spinning speed. The lines on thefigure are linear fits to the data; the slopes of the lines for the

40 INJ Winter 2005

Figure 4TENSILE TESTS RUN ON AGED POLYBUTYLENE FIBERS SPUN USING A FLOWRATE OF 0.5

G/MIN AND SPINNING SPEEDS FROM 500 TO 2500 M/MIN. Final fiber diameters varied from 37.3 microns for a spinning speed of 500 m/min to 16.7 microns for a spinning speed of

2500 m/min. Each curve is the average of from 5 to 8 individual tensile tests. Of the 5-8 fibers involved in these tests, abouthalf were produced during one day’s run, while the other half were produced on another day. There was essentially no dif-

ference between the tensile results for the fibers made on different days.

41 INJ Winter 2005

Figure 5FINAL DIAMETER OF POLYBUTYLENE FIBERS AS A FUNCTION OF SPINNING SPEED

The fibers were spun using polymer flowrates from 0.23 to 1.00 g/min. The points represent the experimental data from off-line diameter measurements, while the lines represent the theoretical diameters for a given polymer flowrate (see the key on

the graph). Equation (3) was used for this calculation, and the assumed polymer density was 0.915 g/cm3.

Figure 6ATENACITY OF AGED POLYBUTYLENE FIBERS AS A FUNCTION OF SPINNING SPEED.

The fibers were spun using polymer flowrates from 0.23 to 1.00 g/min. The lines represent linear fits to the data for therespective polymer flowrates.

different polymer flow rates have almost the same slope, andthere is a significant increase in modulus as the polymer flowrate decreases. Similar to the above-mentioned results fortenacity, polybutylene fibers show a significantly higher mod-ulus than (as spun) polypropylene fibers, especially at verylow polymer flowrates.

To compare the moduli of aged fibers with the moduli offresh fibers, the best-fit lines from Figure 7a were replotted onFigure 7b. Also on Figure 7b are data for the moduli of freshfibers produced at polymer flow rates of 0.5 and 1.0 g/min.These fresh fiber samples were tested within 4 hours after thefibers were melt spun. The data for aged samples in Figure 7awere produced by setting aside fibers from the same batchesof fibers that were used for fresh samples in Figure 7b. Theaged samples were allowed to age at room temperature forone week before the tenacity tests were conducted. Figure 7bshows that the fresh and aged samples have very similarmoduli. Statistically, there are no differences in the moduli.

Fiber Elongation at Break (Eb )The Eb (elongation at break) of polymer fiber is also quite

important. The Eb values for the aged polybutylene fibers areshown in Figure 8a. The Eb decreases rapidly as spinningspeed rises to about 1000 m/min. As speed rises further, Ebbecomes nearly constant, which indicates that most structuraldevelopment in the fiber has been achieved when spinning isat the 1000 m/min level (or above). The final Eb is achieved

earlier for the lower polymer flow rates. To compare the Eb of aged fibers with the Eb of fresh fibers,

the best-fit lines from Figure 8a were replotted on Figure 8b.Also on Figure 8b are data for the Eb of fresh fibers producedat polymer flow rates of 0.5 and 1.0 g/min. This replottingprocedure parallels what was done for Figures 6a and 6b andFigures 7a and 7b. Statistically, there are no differences in the Ebof the aged versus the fresh fibers.

Fiber BirefringenceBirefringence measurements can help determine the orien-

tation in polymer fibers. Figure 9a shows the birefringence ofaged individual filaments of PB-1. Figure 9b shows the bire-fringence results for fresh filaments. In parallel to what wasshown previously for tenacity, modulus, and Eb, there is nostatistical difference in the birefringence values of the freshversus the aged fibers. Similar to the behavior of polypropy-lene and polyester (Vassilatos et al., 1985) the birefringenceincreases as take up velocity increases. Also, the birefringenceincreases as polymer throughput decreases. These resultsparallel the tenacity results shown on Figures 6a and 6b. Thisis expected since, as birefringence increases, orientation andtenacity increase.

Properties of Polybutylene versus Polypropylene The tenacity and elongation at break of polybutylene fibers

are quite different than the same properties for polypropylene

42 INJ Winter 2005

Figure 6BCOMPARISON OF TENACITY OF FRESH POLYBUTYLENE FIBERS

WITH THE TENACITY OF AGED FIBERSThe data points represent fresh fiber tenacities, while the lines represent averages of aged fiber tenacities. The data points

used to produce these lines are shown in Figure 6a.

43 INJ Winter 2005

Figure 7AMODULUS OF AGED POLYBUTYLENE FIBERS AS A FUNCTION OF SPINNING SPEED

The fibers were spun using polymer flowrates from 0.23 to 1.00 g/min. The lines are linear fits to the data.

Figure 7BCOMPARISON OF MODULUS OF FRESH POLYBUTYLENE FIBERS

WITH THE MODULUS OF AGED FIBERSThe data points represent fresh fiber moduli, while the lines represent averages of aged fiber moduli. The data points used

to produce these lines are shown in Figure 7a.

44 INJ Winter 2005

Figure 8AELONGATION AT BREAK (EB) OF AGED POLYBUTYLENE FIBERS

AS A FUNCTION OF SPINNING SPEED. The fibers were spun using polymer flowrates from 0.23 to 1.00 g/min. The data points for the 0.50 and 1.00 g/min

flowrates were fitted using an exponential decay function, while the data for the 0.23 and 0.30 flowrates were fitted with alinear function.

Figure 8BCOMPARISON OF EB OF FRESH POLYBUTYLENE FIBERS WITH THE EB OF AGED FIBERS.

The data points represent fresh fiber Eb, while the lines represent averages of aged fiber Eb. The data points used to produce these lines are shown in Figure 8a.

45 INJ Winter 2005

Figure 9ABIREFRINGENCE OF AGED POLYBUTYLENE FIBERS AS A FUNCTION OF SPINNING SPEED

The fibers were spun using polymer flowrates from 0.23 to 1.00 g/min. The lines are linear fits to the data for each polymer flowrate.

Figure 9BBIREFRINGENCE OF FRESH POLYBUTYLENE FIBERS AS A FUNCTION OF SPINNING SPEEDThe data points represent the birefringence of fresh fibers, while the lines represent averages of the birefringence of aged

fibers. The data points used to produce these lines are shown in Figure 9a.

fibers spun using nearly the same operating conditions. Table3 gives such a comparison. For the values given in the table,the PP was spun at 195 OC, while the PB-1 was spun at 190 OC.All other parameters, such as spinning speeds and polymerthroughputs, were the same for the spinning of both fibertypes. The results for PB-1 are given for the aged fibers. Table3 compares the properties of PB-1 and PP fibers for spinningspeeds of 800-1750 m/min and for two polymer flow rates.

As shown in Table 3, the tenacities of the PB-1 are 3-4 timesthe tenacities of the corresponding PP fibers. An even biggerdifference occurs for the elongation at break (Eb); the Eb forthe PP is 10-30 times larger than the Eb for the PB-1. There isa tremendous amount of draw left in the PP fibers. Of course,at much higher spinning speeds (e.g., at 6000 m/min), the as-spun PP will exhibit stress-strain behavior closer to the PB-1(spun at lower speeds); see Bansal and Shambaugh (1998) fordata on high-speed spinning of PP. Polyester exhibits behav-ior similar to PP. At commercial speeds of 3,200 m/min, poly-ester has residual draw, and the polyester is referred to asPOY (partially oriented yarn). At speeds of 6,400 m/min, thepolyester has almost no remaining draw, and the polyester isoften called “hard” yarn; see Vassilatos et al. (1985). What isinteresting in our present work is that the PB-1 exhibits“hard” properties at much lower spinning speeds than need-ed for PP or polyester. This means that nonwoven webs ofPB-1 could be much stronger than webs made from the com-mercially-common PP. For the case of melt blown fibers, PPwebs often have to be reinforced with spunbonded fibers; thecommon SMS (spunbond-meltblown-spunbond) composite isan example of such a structure.

DMA Results for Aged, Fresh, Solid, and Hollow FibersDynamic mechanical analysis (DMA) was used to examine

our fibers. Fresh fibers were compared to aged fibers, and, inaddition, solid fibers were compared to hollow fibers.Duplicate DMA measurements were carried out on each typeof fiber that was tested. DMA on the fresh fibers was per-formed within 6 hours from the melt spinning process. Theaged fibers were set aside for one week before testing. Forthese DMA tests, all fibers were spun using 1.00 g/min poly-mer flowrate and a spinning speed of 1500 m/min. The solidfibers produced with these conditions had an average diame-ter of 28.9 microns. A nitrogen flowrate of 1.00 ml/min wasused for the production of hollow fibers with approximately50% hollowness. The hollow fibers had an average insidediameter of 26.5 microns and an average outside diameter of38.3 microns, with a standard deviation on the measurementsof about 1.8 microns. Within this standard deviation, thecross-sectional area of hollow fibers is equivalent to the cross-sectional area of the solid fibers. The hollowness of the fiberswas calculated from the formula (de Rovere and Shambaugh,2001)

(4)

where OD is the outer diameter of the fiber and ID is theinner diameter of the fiber.

Figure 10 shows the results from the DMA analysis of fresh,solid PB-1 fibers, and Figure 11 shows the DMA results for theaged, solid samples. The graphs show the storage modulus

46 INJ Winter 2005

Table 3TENACITY AND STANDARD DEVIATION FOR FRESH AND AGED PB-1 FIBERS SPUN

AT 0.5 G/MIN AND DIFFERENT SPINNING SPEEDS.

100%2

2

⋅=OD

IDhollowness

47 INJ Winter 2005

Figure 10DMA RESULTS FOR SOLID, FRESH POLYBUTYLENE FIBERS SPUN USING A POLYMER

FLOWRATE OF 1.00 G/MIN AND A SPINNING SPEED OF 1500 M/MINThe final fiber diameter was 28.9 microns. The DMA was run 6 hours after the fibers were produced.

Two replicate measurements of fibers from the same batch are shown.

Figure 11DMA RESULTS FOR SOLID, AGED POLYBUTYLENE FIBERS SPUN USING A POLYMER

FLOWRATE OF 1.00 G/MIN AND A SPINNING SPEED OF 1500 M/MINThe final fiber diameter was 28.9 microns. The fibers were allowed to age for one week at room temperature before the

DMA was run. Two replicate measurements of fibers from the same batch are shown.

(E’) and the loss modulus (E”) that are calculatedfrom the response of the polymer fiber when subject-ed to a sinusoidal loading. The E’ represents the stiff-ness of the material, and E” represents the energy thatthe fiber loses as a consequence of internal motions orrearrangements. Tan delta is also shown in thegraphs; tan delta is the ratio of the loss modulus tothe elastic modulus. Tan delta represents the efficien-cy of the material with respect to loss and storage ofenergy. As mentioned earlier, the data were taken at1.5 OC increments for the range -40 to 0 OC, while 4 OCincrements were used elsewhere in the temperaturerange (-100 to 120 OC). Both Figure 10 and Figure 11show two replicate measurements on fibers from thesame batch (i.e., fibers spun under the same condi-tions on the same day); as can be observed, there is lit-tle variance between the results for the replicate tests. To compare Figure 10 with Figure 11 (i.e., to comparesolid, fresh with solid, aged fibers), values of E’, E’’,and tan delta were selected for the following threespecific temperatures: -20 OC, 26 OC, and 103 OC.These three temperatures correspond to, respectively,the glass transition temperature of -20 OC (see peakvalue of tan δ on Figure 10), room temperature, and arelatively high use temperature. Table 4 shows thedata for the comparison of Figure 10 with Figure 11.

Solid, fresh samples represent what the manufac-

48 INJ Winter 2005

Figure 12DMA RESULTS FOR HOLLOW, FRESH POLYBUTYLENE FIBERS SPUN USING A POLYMER

FLOWRATE OF 1.00 G/MIN, A NITROGEN FLOWRATE OF 1.00 ML/MIN, AND A SPINNINGSPEED OF 1500 M/MIN

The outside fiber diameter was 38.3 microns, and the inside fiber diameter was 26.5 microns. The DMA was run 6 hoursafter the fibers were produced. Two replicate measurements of fibers from the same batch are shown.

Table 4DMA RESULTS FOR SOLID, FRESH AND SOLID,

AGED POLYBUTYLENE FIBERSThe values shown are averages of the two replicate runs taken

from Figures 10 and 11. The percent difference values were deter-mined from the relation

% difference = (fresh value - aged value)/(aged value)

49 INJ Winter 2005

Figure 13DMA RESULTS FOR HOLLOW, AGED POLYBUTYLENE FIBERS SPUN USING A FLOWRATE OF

1.00 G/MIN, A NITROGEN FLOWRATE OF 1.00 ML/MIN, AND A SPINNING SPEED OF 1500 M/MIN

The outside fiber diameter was 38.3 microns, and the inside fiber diameter was 26.5 microns. The fibers were allowed to age forone week at room temperature before the DMA was run. Two replicate measurements of fibers from the same batch are shown.

Figure 14COMPRESSION AND RECOVERY BEHAVIOR OF SOLID, FRESH POLYBUTYLENE FIBERS SPUN

AT 1500 M/MIN AND POLYMER FLOWRATES OF 0.50, 1.00 AND 2.00 G/MIN. The diameter of the fibers was 21.7 microns, 28.9 microns, and 43.4 microns for polymer flowrates of 0.50 g/min, 1.00

g/min, and 2.00 g/min respectively. Each data point represents the average of eight replicate samples.

turer handles in the plant, while solid, aged samples representwhat the customer receives (assuming that at least two weekspasses between production and use). As it can be seen in Table

4, at the three different temperatures the solid, freshsamples have 28.9 to 54.0 percent higher elastic mod-uli than the solid, one-week-aged fibers. As the tem-perature increases the E’ difference between fresh andaged samples also increases: in the table the loss mod-ulus ranges from 25.9 to 73.0 percent higher for thefresh versus the aged. Also, as with E’, the E” differ-ence is least at the glass transition temperature.

Figure 12 shows the results from the DMA analysisof fresh, hollow PB-1 fibers, and Figure 13 shows theDMA results for the aged, hollow fibers. Both Figure12 and Figure 13 show the data from two replicatemeasurements of fibers from the same batch; as canbe observed, there is little variance between theresults for the replicate tests. To compare Figure 12with Figure 13 (i.e., to compare hollow, fresh withhollow, aged fibers), values of E’, E’’, and tan deltawere selected at temperatures of -20 OC, 26 OC, and 103OC. These selected values are listed in Table 5. As canbe seen, the fresh fibers have a higher E’ and E” thanthe aged fibers, while the tan delta is about the samefor the fibers. The percent difference between the E’and E” properties of the hollow, fresh and hollow,aged fibers (Table 5) is about the same as the differ-ence between the solid, fresh and solid, aged fibers(Table 4).

It is also of interest to compare hollow fibers withsolid fibers. Table 6 gives this comparison. The tableillustrates that, whether the fibers are fresh or aged,the hollow fibers have higher E’ and E” than the solid

fibers. Keep in mind that the cross-sectional area (i.e., thedenier) of the hollow fibers is equivalent to that of the solidfibers. The differences between E’ and E” for hollow versus

solid fibers may be due to the differences in fiberstress during the formation process of the hollow ver-sus the solid fibers.

Compression/Recovery TestsAs described in the experimental details section, we

used a modified INDA test to perform compres-sion/recovery tests on mats made from the PB-1fibers. Both solid and hollow PB-1 fibers were spunat 0.5, 1.0, and 2.0 g/min polymer flow rates and aspinning speed of 1500 m/min. These fibers werecollected as mats. The apparatus and technique usedto analyze the mats is given in de Rovere andShambaugh (2001). The mat thickness was 20-24 mmfor the range of polymer flowrates (0.5, 1.0, and 2.0g/min).

Figure 14 shows the compression/recovery behav-ior for solid, fresh polybutylene fibers. As describedin the procedure developed by de Rovere andShambaugh (2001), the ordinate values show theheight of the fiber web as the stainless steel weight isadded and removed over a 24 hour period. Each datapoint in Figure 14 represents the average of eightreplicate samples. Visual examination of Figure 14

50 INJ Winter 2005

Table 5DMA RESULTS FOR HOLLOW, FRESH AND HOLLOW, AGED POLYBUTYLENE FIBERS.

The values shown are averages of the two replicate runs takenfrom Figures 10 and 11. The percent difference values were deter-

mined from the relation% difference = (fresh value - aged value)/(aged value)

Table 6DMA COMPARISON OF HOLLOW VERSUS SOLID

POLYBUTYLENE FIBERSThe percent difference values were determined from the relation

% difference = (fresh value - aged value)/(aged value)

indicates that using a lower polymer flow rate gives a fibermat with superior recovery. This is an expected result, since

the fibers spun at lower polymer rate – andthe same spinning speed – have better struc-ture development (see Figures 6a, 7a, 8a, and9a).

Point A on Figure 14 represents the initialheight of the stack with the cover plate ontop of the sample but with no weight added.Point B is the height of the stack with theweight added at time zero. Point C is theheight of the stack after 10 min with theweight still upon the stack. Point D is theheight at 10 minutes after the weight isimmediately removed. This procedure ofadding the weight for a time, and thenremoving the weight for a time, is carried onfor 25 hours. Calculated parameters fromthe compression test are defined as follows:

Figure 15 shows these calculated parametersfor the data given in Figure 14. Figure 15reveals that, at the highest polymer flowrate, there is a decrease in compression resis-tance, immediate recovery, and long-termrecovery. The elastic loss is significantlyhigher at the 2.0 g/min flow rate in compar-ison with the other flow rates. There is onlya slight difference between the compressionresults for the 0.5 and 1.0 g/min polymerrates.

For a comparison of fresh versus agedmaterial, compression studies were done onsolid fibers that were fresh, aged 1-week andaged 2-weeks. The samples used for thiscomparison were run at 1.0 g/min polymerflowrate. Figure 16 compares the compres-sion behavior of these fiber mats. The resultsshow that there is no significant difference inthe compression behavior of fresh versusaged PB-1 fibers.

A study on mats of hollow fibers was alsoconducted to compare the fresh with theaged fibers. The hollow fibers were spun at

the same polymer flowrate (1.00 g/min) and the same spin-ning rate (1500 m/min) that were used to produce the solid

51 INJ Winter 2005

Figure 15COMPRESSION AND RECOVERY CALCULATED PARAME-

TERS FOR SOLID, FRESH POLYBUTYLENE FIBERS SPUNUSING POLYMER FLOWRATES OF 0.50, 1.00 AND 2.00

G/MIN, AND A SPINNING SPEED OF 1500 M/MINThe diameter of the fibers varied from 21.7 microns for the 0.50 g/min flowrate

to 43.4 microns for the polymer flowrate of 2.00 g/min.

Figure 16COMPRESSION AND RECOVERY PARAMETERS FOR FRESH,

ONE-WEEK AGED, AND TWO-WEEK AGED SOLID POLYBUTYLENE FIBERS

The fibers were spun using a polymer flowrate of 1.00 g/min and a spinningspeed of 1500 m/min. The fiber diameter was 28.9 microns.

fibers examined in Figures 15 and 16. To produce about 50%hollowness, a nitrogen flowrate of 1.00 ml/min was used.Figure 17 shows the compression behavior for mats of thesehollow PB-1 fibers. Mats of the fresh fibers have a little high-er elastic loss, but the other three parameters are essentially

the same for the fresh versus the aged sam-ples.

The data in Figures 16 and 17 can be usedto compare mats of hollow fibers with matsof solid fibers. Table 7 shows this compari-son. Hollow PB-1 fibers exhibit slightly lesscompression resistance and slightly greaterelastic loss than the solid filaments. TheImmediate recovery and the long-termrecovery are essentially the same for the hol-low and the solid fibers. In past work by deRovere and Shambaugh (2001), it was foundthat mats of hollow PP fibers have essential-ly the same compression/recovery parame-ters as do mats of solid PP fibers. Data fromde Rovere and Shambaugh are included inTable 7. Their PP mats have essentially thesame immediate recovery as our polybuty-lene mats. However, their PP mats had lesscompression resistance and less long-termrecovery, and their mats had higher elasticloss.

ConclusionsMelt-spun polybutylene fibers have high

moduli and high tenacities compared topolypropylene or polyester fibers spun atsimilar speeds (250-2500 m/min). Hence,

polybutylene fibers may be useful in producing strong websof nonwoven fibers in processes that can capitalize on thesehigh moduli and tenacities.

The properties of the polybutylene fibers and poly-butylene nonwoven mats change relatively little with aging.

52 INJ Winter 2005

Figure 17COMPRESSION AND RECOVERY PARAMETERS FOR FRESH,

ONE-WEEK AGED, AND TWO-WEEK AGED HOLLOW POLYBUTYLENE FIBERS.

The fibers were spun using a polymer flowrate of 1.00 g/min, a nitrogenflowrate of 1.00 mL/min, and a spinning speed of 1500 m/min. The fibers hadan outside diameter of 38.3 microns, and inside diameter of 26.5 microns, and a

hollowness of 50%.

Table 7COMPARISON OF MATS OF HOLLOW POLYBUTYLENE FIBERS WITH MATS OF SOLID

POLYBUTYLENE FIBERS. All fibers were made at a spinning speed of 1500 m/min and a polymer flowrate of 1.00 g/min (so the denier of all

fibers was the same). The hollow fibers were made with a 1.00 ml/min nitrogen rate, which gave a hollowness of about50%. The polybutylene data are from Figures 16 and 17. The PP data are from de Rovere and Shambaugh (2001) and

are values for 51% hollow fibers with an OD of 47 μm.

Hence, aging should not be a problem in the use of thesefibers. Nonwoven mats of polybutylene havecompression/recovery properties that are as good as the com-pression/recovery properties of common polypropylenemats.

DMA results for hollow polybutylene fibers show that thehollow fibers have higher E’ and E” than the solid fibers.However, compression/recovery of mats of hollow fibers isessentially the same as the compression/recovery of mats ofsolid fibers.

AcknowledgementsThe authors are grateful for the financial support provided

by 3M, Procter & Gamble, and Du Pont.

NomenclatureA= initial height of stack with cover plate on sample but no

additional weightB= height of the stack with the weight added at time zero C= height of the stack after 10 min with the weight addedD= height of the stack after 10 min with weight removedE= height of the stack after 20 min with weight removedF= height of the stack after 20 min with the weight addedG= height of the stack after 30 min with the weight addedH= height of the stack after 30 min with weight removedI= height of the stack after 40 min with weight removedJ= height of the stack after 40 min with the weight addedK= height of the stack after 50 min with the weight addedL= height of the stack after 50 min with weight removedM= height of the stack after 60 min with weight removedN= height of the stack after 8 hours with weight removedO= height of the stack after 8 hours with the weight addedP= height of the stack after 24 hours with the weight addedQ= height of the stack after 24 hours with weight removedR= height of the stack after 25 hours with weight removed

d = fiber diameter, nmE’= modulus of elasticity, dyn/cm2E” = loss modulus, dyn/cm2H = fiber hollownessID = fiber inner diameterm = mass flow rate, g/minnparallel = refractive index parallel to the fiber axisnperpendicular = refractive index perpendicular to the fiber

axisOD = fiber outer diameterR = relative retardation, nmv = fiber velocity, m/min

Greek SymbolsΔn = fiber birefringenceρ = fiber density, g/cm3tan δ = damping factor

ReferencesAhmed, M. Polypropylene Fibers- Science and Technology;

Elsevier Scientific Publishing Company: New York, 1982.

Bansal, V.; Shambaugh, R.L. "On-line Density andCrystallinity of Polyethylene Terephthalate During MeltSpinning", Polym. Eng. Sci., 1998, 38(12), 1959-1968.

Bansal, V.; Shambaugh, R.L. "On-line Determination ofDiameter and Temperature During Melt Blowing ofPolypropylene", Ind. Eng. Chem. Res., 1998, 37(5), 1799-1806.

Boor, J.; Mitchell, J. C. "Kinetics of crystallization and crys-tal-crystal transition in poly-1-butylene", J. Polym. Sci. 1963,A1, 59.

Choi, C. H.; White, J. L. "Comparative Study of StructureDevelopment in Melt Spinning Polyolefin Fibers",International Polymer Processing XIII, 1998, 13 (1), 78-87.

Choi, C. H.; White, J. L. "Crystal-Crystal Transformations inIsotactic Polybutene-1 Oriented Filaments and in thickMolded Rods", Polym. Eng. Sci. 2001, 41 (6), 933-939.

Choi, S. Y.; Rakus, J. P.; O’Toole, J. L. "Time DependentCrystallinity of Polybutene-1", Polym. Eng. Sci. 1966, 6 (4), 349-352.

de Rovere, A.; Grady, B.; Shambaugh, R. L. "The Influenceof Processing Parameters on the Properties of Melt-SpunPolypropylene Hollow Fibers", J. Appl. Polym. Sci. 2002, 83,1759-1772.

de Rovere, A.; Shambaugh, R. L. "Melt-spun Hollow Fibersfor Use in Non-woven Structures", Ind. Eng. Chem. Res. 2001,40 (1), 176-187.

Holland, V. F.; Miller, R. L. "Isotactic Polybutene-1 SingleCrystals: Morphology", J. Appl. Phys. 1964, 35 (11), 3241-3248.

Katayama, K.; Amano, T., Kolloid Z.-Z Polym. 1968, 226,125.

Miller, R. L.; Nielsen, L. E. “Crystallographic Data forVarious Polymers”, Appendix I in “Polymer Reviews, Vol 5:Polymer Single Crystals”, Wiley-Interscience, New York,1963.

Nakamura, K.; Aoike, T.; Usaka, K.; Kanamoto, T. “PhaseTransformation in Poly(1-butene) upon Drawing”,Macromolecules, 1999, 32, 4975-4982.

Natta, G.; Corradini, P. "The Crystal Structure of IsotacticPoly-1-butene", Makromo. Chem. 1956, 21, 240-244.

Natta, G.; Corradini, P.; Bassi, I. W. "Crystal Structure ofIsotactic Poly(1-Butene)", Nuovo Cimento 1960, 15 (1), 52-67.

Nonwovens.com, http://www.nonwovens.com/market/regional/na.shtml, 2004.

Phillips, W. R. “Mineral Optics: Principles and Techniques”,W. H. Freeman and Company, San Francisco, 1971, pp. 92-98

Rohn, C. L. "Predicting the Properties of Poly-(Butene-1)Blown Films", J. Polym. Sci.: Symposium 1974, 46, 161-174.

Samon, J. M.; Schultz, J. M.; Hsiao, B. S.; Wu, J.; Khot, S."Structure Development during Melt Spinning andSubsequent Annealing of Polybutene-1 Fibers", J. Polym. Sci.:Part B: Polym. Phys. 2000, 38, 1872-1882.

Sasaguri, K.; Rhodes, M. B.; Stein, R. S. "Spherulite defor-mation in poly-1- butene", J. Polym. Sci. 1963, B1 (1), 571-574.

Schemm, F.; Grasmeder, J.; Konnecke, K.; Van de Vliet, F."Polybutene-1 Piping Systems: Material-specific Propertiesand Typical Applications", Proceedings of WiesbadenerKunststoffrohrtage Conference, Wiesbaden, Germany, April25-26 2002.

53 INJ Winter 2005

Shaw, J. P.; Gilbert, M. "Characterization of CommercialPolybutylenes. Part II. Crystallinity and Tensile Properties", J.Macromol. Sci. Phys. 1991, B30 (4), 301-317.

Shaw, J. P.; Gilbert, M. "Polybutylene Monofilaments. I:Preparation and Properties", Plastics, Rubber and CompositesProcessing and Applications. 1994, 22, 9-27.

Standard Method for Determination of Compression andRecovery of Highloft Nonwoven Fabrics at RoomTemperature Using Weights and Plates; INDA IST 120.4 (95),Association of the Nonwoven Fabrics Industry: Cary, NC,1995

Vassilatos, G.; Knox, B.H.; Frankfort, H.R.E. “Dynamics,Structure Development, and Fiber Properties in High-SpeedSpinning of Polyethylene Terephthalate”, in “High-SpeedFiber Spinning”, edited by A. Ziabicki, and H. Kawai, Wiley-Interscience, New York, 1985, pp. 384, 389, and 390.

Weynant, E.; Haudin, J. M.; G'Sell, C. "Plastic deformationand solid-phase transformation in polybutene-1", J. Mat. Sci.1982, 17, 1017.

Zanetti, R.; Manaresi, P.; Buzzoni, G. C. "Crystallinity andPolymorphism of Poly-a-butene", Chimica e l'Industria 1961,43, 735-740. — INJ

54 INJ Winter 2005

The International Nonwovens Journal is brought to you fromAssociations from around the world. This critical technicalpublication is provided as a complimentary service to the

membership of the Associations that providedthe funding and hard work.

PUBLISHER

INDA, ASSOCIATION OF THE NONWOVEN FABRICS INDUSTRYRORY HOLMES

PRESIDENTP.O. BOX 1288, CARY, NC 27511

www.inda.org

SPONSOR

TAPPI, TECHNICAL ASSOCIATION OF THE PULP AND PAPER INDUSTRYWAYNE H. GROSS

EXECUTIVE DIRECTOR/COOP.O. BOX 105113

ATLANTA, GA 30348-5113www.tappi.org