Introduction

Competition among high-performance fibers has prompted new processing methods for obtaining high-performance fibers of ultrahigh molecular weight polyethylene (UHMWPE) [1–11]. The key element in obtaining high-strength UHMWPE fibers is to find a way to draw the as-prepared gel specimens to an ultrahigh draw ratio after the gel spinning process. The drawability of the as-prepared gel specimens was found to depend significantly on the compositions of solutions from which gels were made [12–17]. Several authors [16, 18–22] reported that the drawing temperature and rate could markedly affect the maximal achievable draw ratio and tensile properties of solution-grown UHMWPE samples. In addition to the gel solution compositions and drawing conditions, it is generally recognized that the conditions used in the formation process after spinning and/or solution casting of gel solutions can also have a significant influence on the morphology, microstructure and drawing properties of the specimens formed during the above mentioned processes [2, 19, 23–30].

The discovery of carbon nanotubes (CNTs) has attracted enormous attention in recent years due to their remarkable physical characteristics, such as, extremely high Young’s modulus and rupture strength, high aspect ratio, nanometer-range diameter [25]. Presumably, incorporation of uniformly dispersed and aligned CNTs in polymer matrix can provide polymer composites with dramatically improved strength and modulus in their machine direction. These expectations have recently been confirmed by a number of studies [26–35]. Our recent investigation [32–34] found that the achievable draw ratios (Dra) of UHMWPE/carbon nanotubes (CNTs) as-prepared fibers prepared near the optimal UHMWPE concentration improve consistently and reach a maximal value as their CNT and/or functionalized CNT contents increase up to an optimal value, respectively. As evidenced by DSC thermal analysis, CNTs with extremely high specific surface areas can serve as efficient nucleation sites and facilitate the crystallization of UHMWPE molecules into crystals but with low melting temperatures and/or evaluated smaller crystal thickness during their crystallization processes. Presumably, the crystals with lower melting temperatures and/or evaluated smaller crystal thickness obtained at proper plain and/or functionalized CNT contents can be melted and pulled out of folded lamellar crystals relatively easily during ultradrawing processes, and hence, results in higher drawability and orientation of the UHMWPE/CNTs or UHMWPE/functionalized CNTs as-prepared fibers. The ultimate tensile strength values of the UHMWPE/CNTs or UHMWPE/functionalized CNTs drawn fibers prepared using one-stage drawing process at 95 °C can reach 2.5 and 3.5 GN m −2, respectively, which is about 17 and 66 % higher than that of the corresponding plain UHMWPE drawn fibers prepared at the same optimal UHMWPE concentration and drawing condition but without CNTs.

Similar to CNTs, ATP is a natural clay mineral together with sepiolite that forms the group of fibrous clay minerals with extremely high specific surface areas (≧ 200 m 2 g −1) [36–40]. Its structure scheme was firstly proposed by Bradly in 1940 [37], who described a theoretical formula of (Mg, Al, Fe)5Si8O20(Al)(HO)2(OH2)4 · 4H2O. In our recent study [41], natural and purified attapulgite (ATP) minerals were added in UHMWPE gel-spinning solutions to improve the ultradrawing and ultimate tensile properties of UHMWPE/ATPs fibers. Similar to those found for UHMWPE/CNTs or UHMWPE/functionalized CNTs as-prepared fibers, the achievable draw ratios of the UHMWPE/ATPs (FAx) and UHMWPE/purified ATPs (FAmx) as-prepared fibers and tensile strength values of their drawn fibers approach a maximal value as their ATPs and/or purified ATPs contents reach the optimal values at 0.02 and 0.015 wt%, respectively. The maximal achievable draw ratios of FAmx as-prepared series fiber specimens and the tensile strengths of the drawn FAmx fiber specimens are significantly higher than those of the corresponding FAx drawn fiber specimens prepared at the same draw ratios and ATPs contents but without purification by sodium hexametaphosphate. The ultimate tensile strength values of the UHMWPE/purified ATPs drawn fibers prepared at the optimal ATP content using one-stage drawing process at 95 °C can reach around 3.9 GN m −2, which is even higher than most of those UHMWPE/functionalized CNTs drawn fibers prepared at the same drawing condition in our previous study [32–34, 41]. These results clearly suggest that nanofillers with extremely high specific surface areas can act as efficient nucleation sites and significantly improve the achievable draw ratios and ultimate tenacity properties of nanofillers added UHMWPE fibers.

In order to further improve the ultradrawing and ultimate tenacity properties of ATP added UHMWPE fibers, ATP nanofibers were etched by hydrochloric acid with varying concentrations for varying amounts of time to increase their specific surface areas in this study. The maximal achievable draw ratio obtained for each UHMWPE/acid treated ATP as-prepared fiber series specimens was significantly higher that of the UHMWPE/purified ATP as-prepared fiber specimens prepared with the same optimal ATP content but without acid treatment. In which, the specific surface areas of ATP nanofibers were found to have a beneficial effect on the ultradrawing and ultimate tenacity properties of UHMWPE/acid treated ATP fibers. In order to understand the significantly improved ultradrawing and ultimate tenacity properties of the UHMWPE/acid treated ATP fibers, elemental, morphological and Fourier transform infrared analyses of the purified and acid treated ATP specimens and/or thermal, orientation factor, morphological and tensile properties of the as-prepared and drawn UHMWPE/purified ATP and UHMWPE/acid treated ATP fiber specimens were investigated.

Experimental

Materials and sample preparation

The UHMWPE resin and attapulgites (ATPs) used in this study were kindly supplied by Yung Chia Chemical Industrial Corporation, Kaohsiung, Taiwan and Xinyuan Science and Technology Corporation, Jiangsu, China, respectively, in which UHMWPE resin is associated with a weight-average molecular weight (Mw) of 5.0 × 106. The ATP ore was first crushed by a hammer and then ground by a planetary ball mill (QM-ISPO4), which was manufactured by Nanjing University Instrument Plant, Nanjing, China. The ground ATP nanofibers were firstly purified in sodium hexametaphosphate (SHMP) deionized water solutions under ultrasonic and stirring condition at 25 °C for 7 h, in whcih 20 g of ATP nanofibers together with 0.6 g of SHMP were dispersed and dissolved in 0.4 l of deionized water. Demarcated black impurity was found precipitating from the suspended upper layer solutions after setting the ultrasonic solutions still for half an hour. The suspended upper layer solutions were then centrifuged to remove impurities using a centrifuge model KDC-160HR for 15 min. The centrifuged ATP residue was dried at 80 °C for 24 h and then ground into purified ATP nanofibers using the planetary ball mill. Five grams of purified ATP nanofibers were treated with 100 ml of 1, 2, 3, 5 and 7 M hydrochloric acid (HCl) solutions under stirring condition at 70 °C for varying amounts of time, respectively. In which, 70 °C was chosen as the acid-treating temperature and used in the above experiments, since the acid treated ATP specimens treated with 3 M HCl solutions at 70 °C for 60 min exhibit significantly higher specific surface areas than those of acid treated ATP fibers treated with other temperatures for 60 min in our preliminary experiments. After treatment of HCl solutions, the ATP containing suspensions were filtered and washed with deionized water to neutral. The HCl treated ATP nanofibers were then dried at 80 °C for 24 h.

Varying compositions of purified or acid treated ATPs nanofibers together with UHMWPE resin were then dispersed and/or dissolved in decalin solvent at 135 °C for 1.5 h, in which 0.1 wt% of di-t-butyl-p-cresol was added as an antioxidant. The original and/or acid treated ATP nanofibers was dried at 80 °C for 24 h in an vacuum oven, before incorporation into the UHMWPE/ATP and/or UHMWPE/acid treated ATP gel solutions. The UHMWPE/ATP and/or UHMWPE/acid treated ATP gel solutions prepared above were then fed into a temperature-controlled hopper and kept as hot homogenized gel solutions before spinning. The hot homogenized gel solutions were then gel-spun using a conical die with an exit diameter of 1 mm at an extrusion rate of 1,000 mm/min and an extrusion temperature of 170 °C. A water bath and a winder with 70 mm in diameter were placed at a distance of 520 mm and 810 mm from the spinneret exit, respectively. The extruded gel fibers were cooled in a temperature-conditioned atmosphere and then quenched into a water bath for about 1 min, where the temperatures of the air atmosphere and water bath were controlled at 5 °C. The quenched fibers were then extracted in a n-hexane bath for 5 min to remove the residual decalin solvent. The extracted fiber specimens were then dried in air for 30 min to remove the remaining n-hexane solvent before any drawing run. Tables 1 and 2.

Table 1 Designations of purified and acid treated ATP specimens prepared in this study
Full size table
Table 2 Designations and compositions of UHMWPE, UHMWPE/ATP and UHMWPE/acid treated ATP as-prepared fiber specimens
Full size table

Morphological and elemental analysis

In order to understand the morphologies and elemental compositions on the surfaces of the purified and acid treated ATP nanofiber specimens prepared in “Materials and sample preparation” section. The original, purified and acid treated ATP nanofibers were dried at 80 °C for 24 h and then gold-coated at 15 kev for 10 s. The original, purified and acid treated ATP nanofibers were also dried and dispersed in alcohol and then dried onto a carbon-coated copper grid under ambient conditions prior to morphological and elemental analysis. A special accessory kit (attenuated total reflectance (ATR)) was used in elemental analysis to determine the content of light elements such as oxygen in original and/or acid treated ATP nanofibers. The UHMWPE, UHMWPE/purified ATP and UHMWPE/acid treated ATP as-prepared and drawn gel fibers were etched with excess amount of fuming nitric acid to highlight the crystalline morphology of the fiber specimen. The amorphous regions of fiber specimens could be etched with ultrasonic generator at 300 W at 60 °C. After the 6 h acid treatment the sample was washed subsequently by both de-ionized water by boiling acetone, and then dried at room temperature. The dried etched fiber specimens were then gold-coated at 15 kev for 10 s. The gold-coated ATP nanofiber and fiber specimens prepared above were then examined using a JEOL Field Emission Scanning Electron Microscope (FESEM) model JSM-6500 operated at 15 kV.

Specific surface area analysis

A Laser Particle Size Analyzer model BT-9300H (Dandong Bettersize Instruments Ltd., Dandong, China) was used to study the specific surface areas of the purified and acid treated ATP specimens prepared in the “Materials and sample preparation” section, although the specific surface areas of ATP and/or acid treated ATP fibers may be underestimated due to their porosity and surface roughness in feature. Before analysis, ten micrograms of purified and acid treated ATP nanofibers were added and ultrasonicated in 10 ml decalin at 25°C for 5 min, respectively. The specific surface areas of purified and acid treated ATP specimens were then measured by placing the ultrasonicated solutions prepared above in the curette of the Laser Particle Size Analyzer at 25°C.

Fourier transform infrared spectroscopy

Fourier transform infrared (FT-IR) spectroscopic measurements of the purified and acid treated ATP specimens were recorded on a Nicolet Avatar 360 FT-IR spectrophotometer at 25 °C, wherein 32 scans with a spectral resolution 1 cm−1 were collected during each spectroscopic measurement. Infrared spectra of the film specimens were determined using the conventional KBr disk method. The original and modified ATP nanofibers were cast onto KBr disk and dried at 60 °C for 30 min. The cast films used in this study were prepared sufficiently thin enough to obey the Beer-Lambert law.

Thermal, lamellar thickness and orientation factor analysis

Thermal properties of all samples were performed on a TA differential scanning calorimeter (DSC) model Q100. All scans were carried out at a heating rate of 20°C/min under flowing nitrogen at a flow rate of 25 ml/min. Samples weighing 0.5 mg and 15 mg were placed in the standard aluminum sample pans for determination of their melting temperature (Tm) and percentage crystallinity values, respectively. The percentage crystallinity values of the specimens were estimated using baselines drawn from 40 to 200 °C and a perfect heat of fusion of polyethylene of 293 J/g [42, 43].

In order to understand the ultradrawing properties of UHMWPE, UHMWPE/ATP and UHMWPE/acid treated ATP as-prepared fiber specimens, the lamellar thickness (lc) values of the above as-prepared fibers were evaluated from their Tm values using Hoffman and Week’s equation [43, 44] given as follows. Broad melting peak observed from non-isothermal DSC scans and the effect of lamella thickening may all lead to deviations from the true lc values. Nevertheless, Hoffman-Weeks equation was used for qualitative estimation of lc values of UHMWPE, UHMWPE/ATP and/or UHMWPE/acid treated ATP as-prepared fiber specimens, because it is the most feasible technique one can use to estimate the lc values of fiber specimens. In which, an equilibrium melting temperature (To m) of 145.5 °C, a perfect heat of fusion (ΔHf 0) of 293 J/g and a folded surface free energy (σe) of 9 × 10−6 J/cm2 of polyethylene crystals [43] were used for evaluation of lc values of UHMWPE, UHMWPE/ATP and UHMWPE/acid treated ATP as-prepared fiber specimens.

$$ {\mathrm{T}}_{\mathrm{m}}={\mathrm{T}}_{\mathrm{m}}^{\mathrm{o}}\left[1-\frac{2{\sigma}_{\mathrm{e}}}{1_c\varDelta {\mathrm{H}}_{\mathrm{f}}^0}\right] $$
(1)

The orientation factor (fo) values of UHMWPE, UHMWPE/ATP and UHMWPE/acid treated ATP as-prepared and drawn fiber specimens were measured using a sonic velocity orientation instrument model SCY-III, which was purchased from Donghuakaili Chemicals and Fiber Technology Corporation, Shanghai, China. Before testing, the fiber specimen with 60 cm in length was wound and clamped on a testing device with a span of 40 cm. The fo values of the as-spun and drawn fiber specimens were then measured at 25 °C. A minimum of five samples of each specimen were tested and averaged during the orientation measurements. The f0 values were evaluated using Eq. (2) as suggested by Xiao and coauthors [45]:

$$ {\mathrm{f}}_0=1-{\left({\mathrm{C}}_{\mathrm{u}}/\mathrm{C}\right)}^2 $$
(2)

where C is the sonic velocity of the as-prepared or drawn UHMWPE fiber specimen and Cu is the sonic velocity of the fully unoriented sample, taken as 1.65 km/s [45].

Drawing and tensile properties of the fiber specimens

The UHMWPE, UHMWPE/ATP and UHMWPE/acid treated ATP fiber specimens used in the drawing experiments were cut from the dried as-prepared fibers and then stretched on a Gotech tension testing machine model GT-TFS-2000 equipped with a temperature-controlled oven. The fibers are 20 mm in length, which were wound and clamped in a stretching device and then stretched at 95 °C and a crosshead speed of 20 mm/min. The draw ratio of each fiber specimen was determined as the ratio of the marked displacement after and before drawing. The marked displacement before drawing was 27 mm. The tensile properties of the as-prepared and drawn fiber specimens were also determined using a Hung-Ta tension testing machine model HT-9112 at a crosshead speed of 20 mm/min. A minimum of five samples of each specimen were tested and averaged during the tensile experiments.

Results and Discussion

Morphological analyses of the original, purified and acid treated ATPs

Figure 1 exhibits typical SEM micrographs of original, purified and acid treated ATP specimens. Typical rodlike feature with dimensions of 150–600 nm in length and 10–45 nm in diameter was observed for original and purified ATP specimens. Before purification, the rodlike ATP nanofibers were found agglomerating and/or bundling together significantly on SEM micrograph of the original ATP specimen (see Fig. 1a). As shown in Fig 1b, the agglomerated bundles of ATP nanofibers originally present in original ATP specimen were significantly dispersed after the impurities were removed between ATP nanofibers. After acid treatment for a fixed amount of time, the surface roughness of acid treated ATP nanofibers seemed to increase due to hydrochloric acid etching even at low HCl concentrations, although agglomeration of ATP nanofibers was still present in acid-treated ATP specimens without ultrasonication. Moreover, the rodlike feature of ATP nanofibers were significantly etched into broken/shorter and thinner nanofibers by HCl solutions as their concentrations increase from 1 to 7 M (see Figs. 1c to g). In fact, after etching by 7 M HCl solution for 60 min, nearly only broken/short ATP nanofibers were found on the SEM micrograph of A7M specimen (see Fig. 1g).

Fig. 1
figure 1

SEM micrographs of a ATP, b purified ATP and c A1M, d A2M, e A3M, f A5M and g A7M acid treated specimens treated with HCl solutions for 60 min

Full size image

Figure 2 summarized typical SEM micrographs of A5M specimens treated with 5 M HCl solution for various amounts of time. Similarly, after treating with 5 M HCl solutions, the rodlike feature of ATP nanofibers were etched into broken/shorter and thinner rodlike feature as the treating time increase from 20 to 100 min (see Fig. 2a to e). In fact, nearly only broken and short ATP nanofibers were found on the SEM micrograph of A5M specimens after they were excessive etched with 5 M HCl solutions for more than 60 min (see Fig. 2d to e).

Fig. 2
figure 2

SEM micrographs of acid treated ATP specimens after acid treatment with 5 M HCl solution for a 20, b 40, c 60, d 80, and e 100 min

Full size image

Elemental analyses of the original, purified and acid treated ATPs

Table 3 summarized the compositions of original, purified and acid treated ATP specimens after elemental analyses. Elements, such as Si, Al and Mg were found [46] as the main compositions on the surfaces of original ATPs. The small amounts of P, Ca, Ni and Ti elements found on the surfaces of original ATPs are ascribed as the impurities that present as mineral compounds in the original ATPs, in which Ca element is reported present as CaCO3 containing minerals [47]. After purification, the contents of Ca, Ni and Ti elements reduce significantly to 0.4, 0.6 and 0.3 wt%, respectively; but P contents increase from 0.5 to 4.3 wt%. Moreover, as shown in Table 4, Ca, Ni and Ti elements were indeed found as the main compositions of sediments after purification processes. These results suggest that Ca, Ni and Ti contained impurities originally present on the surfaces of ATP nanofibers precipitated as black sediments after their purification processes. During purification processes of ATP nanofibers, the abundant unbalanced charges present on surfaces of ATPs are likely to adsorb SHMP molecules with high P contents. As a consequence, Ni, Ti and Ca contained impurities reduce but P contained impurities increase significantly after the purification processes of ATP nanofibers.

Table 3 Elemental analyses of original, purified and acid treated ATP specimens
Full size table
Table 4 Elemental analyses of sediments after the purification processes of ATP specimens
Full size table

After acid treatment for 60 min, the contents of P, Ni and Ti elements found on the surfaces of acid treated ATP nanofibers restore gradually to those contents found for the original ATP nanofibers, respectively, as the hydrochloric acid (HCl) concentrations used acid treatment increase from 1 to 7 M. In contrast, the contents of Ca element found on the surfaces of acid treated ATP specimens treated with HCl solutions for 60 min remain at low values 0.2 to 0.4 wt%, which are about the same as those found for the purified ATP specimens. As evidenced by morphological analyses in the previous section, after 60 min of acid treatment, the rodlike feature of ATP nanofibers were significantly etched into broken/shorter and/or thinner rodlike feature by HCl solutions as their concentrations increase from 1 to 7 M. These results clearly suggest that P, Ni and Ti elements originally present in ATP nanofibers gradually exposed after acid treatment for 60 min and P, Ni and Ti contents restored gradually to those found for purified ATP nanofibers, as the HCl concentrations used in the acid treatment increased from 1 to 7 M. The relatively low Ca contents found on the surfaces of acid treated ATP specimens is attributed to the easy dissolution of newly exposed CaCO3 impurities in HCl solutions during the acid treating processes.

Specific surface area analyses

The values of specific surface areas of the purified and acid treated ATP nanofibers are summarized in Fig. 3. The specific surface area of the purified ATPs is relatively high and reaches 205 m2/g. After the treatment of HCl solutions, the specific surface areas of acid treated ATP nanofibers are significantly higher than those of the purified ATP nanofibers. In fact, the specific surface areas of each acid treated ATP nanofibers increase significantly with the increase in treating time, and then reach a maximal value as the treating time approaches an appropriate value around 60 min. The maximal specific surface areas obtained for acid treated ATP specimens reach another maximal value as the concentrations of HCl solutions used for 60 min of acid treatment approach an optimal value at 5 M. As shown in Fig. 3, after treating ATP nanofibers in 5 M HCl solution for 60 min, the specific surface areas of the acid treated ATP nanofibers reaches around 282 m2/g, which is about 40 % higher than the specific surface area of purified ATP nanofibers without acid treatment. At HCl concentrations higher than 5 M or treating time more than the optimal value at 60 min, the specific surface areas of the acid treated ATP nanofibers reduce significantly with the further increase in the HCl concentration or treating time. For instance, the maximal specific surface areas of the ATP nanofibers acid-treated for an optimal amount of time at 60 min reduce from 282 m2/g to 250 m2/g, as the HCl concentrations increase from 5 M to 7 M. Similarly, the specific surface areas of ATP nanofibers treated with 5 M HCl concentration reduce from 282 m2/g to 275 m2/g, as the treating time increases from 60 to 80 min. As evidenced by morphological analysis in the previous section, the significant reduction in specific surface areas of acid treated ATP nanofibers with excessive treating time or HCl concentrations is attributed to the collapse of long ATP nanofibers originally with many free tunnels into significantly thinner and shorter ATP nanafibers with less free tunnels or cavities caused by over-etching of the HCl solutions.

Fig. 3
figure 3

Specific surface areas of A (◆), A1M (■), A2M (▲), A3M (×), A5M (+) and A7M (●) specimens

Full size image

Fourier transform infra-red spectroscopy

Figure 4 illustrates typical Fourier transform infra-red (FT-IR) spectra of original, purified and acid treated ATP specimens. As shown in the FT-IR spectrum of the original ATP specimen, seven distinguished absorption bands were found centering at 793, 882, 1,194, 1,433, 1,655, 3,582 and 3,616 cm−1, which were generally ascribed to the motions of Ti-O bending, carbonate bending, Mg-O stretching, carbonate bending, absorbed water bending, Al-Mg-OH bending and Al2-OH bending, respectively[41, 46]. After purification, the Ti-O and carbonate bending bands nearly disappeared, while demarcated Mg-O stretching, absorbed water bending, Al-Mg-OH bending and Al2-OH bending bands were found centering at 1,195, 1,657, 3,583 and 3,616 cm-1 of the FT-IR spectrum of purified ATP specimen (see Fig. 4b), respectively. Similar to those found for purified ATP specimen, demarcated Mg-O stretching, absorbed water bending, Al-Mg-OH bending and Al2-OH bending bands were found centering at 1,195, 1,657, 3,583 and 3,615 cm−1, respectively, of the FT-IR spectra of acid treated ATP specimen. However, Ti-O bending bands reappeared gradually and became more demarcated as HCl concentrations used for acid treatment increased from 1 to 7 M.

Fig. 4
figure 4

FT-IR spectra of the a original ATP, b A, c A1M, d A2M, e A3M, f A5M and g A7M specimens

Full size image

The disappearance of the Ti-O bending bands of purified ATP specimem is likely due to the adsorption of Ti2+ by SHMP after purification of ATP nanofibers. Whereas, the reappearance of Ti-O bending bands of acid treated ATP specimens is attributed to the newly exposed impurities containing Ti-O groups after acid treatment of purified ATP nanofibers. The disappearance of carbonate bending bands of the purified ATP specimens is most likely due to the adsorption of Ca2+ by SHMP after purification of ATP nanofibers, wherein CO3 2− anions were combined with Na+ cations of SHMP to form Na2CO3 and then dissolved or precipitated in water. In contrast, the disappearance of carbonate bending bands of the acid treated ATP specimens is attributed to the easy dissolution of newly exposed CaCO3 impurities in HCl solutions during the acid treating processes.

Thermal properties of the as-prepared fibers

Typical DSC thermograms of UHMWPE (F100), UHMWPE/purified ATP (F100Ay) and UHMWPE/acid treated ATP (F100AxM-y) as-prepared fiber series specimens are summarized in Figs. 5, 6, 7, 8, 9 and 10. Table 5 summarized the lamellar thickness (lc) values of F100, F100Ay and F100AxM-y as-prepared fiber specimens evaluated from their melting temperature (Tm) values using Hoffman and Week’s equation [43, 44]. A main melting endotherm with a peak melting temperature (Tm) and percentage crystallinity (Xc) at 139.6 °C and 65.3 %, respectively, was found for the F100 as-prepared fiber specimen. The Tm(or evaluated lc) and Xc values of F100Ay and/or F100AxM-y as-prepared fiber series specimens approach the minimal and maximal values, as their purified and/or acid treated ATP approach the optimal values at 0.05 and 0.025 phr, respectively. In which, Tm(or evaluated lc) and Xc values of the F100AxM-y as-prepared fiber specimens are significantly lower and higher than those of the corresponding F100Ay as-prepared fiber specimens, respectively. For instance, Tm, lc and Xc values of F100A5M-0.025/F100A1M-0.05as-prepared fiber specimen are 135.8 °C/137.4 °C, 9.6 nm/11.5 nm and 74.6 %/70.2 %,, respectively, which are significantly lower than Tm and lc values (i.e. 139.6 °C, 15.7 nm), but higher than Xc values (i.e. 65.3 %) of F100 as-prepared fiber specimen. Moreover, it is worth noting that each F100A5M-y as-prepared fiber series specimen exhibits the lowest Tm(or evaluated lc) but the highest Xc values among F100AxM-y as-prepared fiber series specimens with the same contents of acid treated ATP nanofibers treating with different HCl concentrations for 60 min (see Figs. 5, 6, 7, 8, 9 and 10). For instance, the Tm(or evaluated lc) and Xc values of F100A5M-0.025 as-prepared fiber are about 0.3 to 0.8 °C (or 0.3 to 0.9 nm) lower and 0.9 to 3.2 % higher than those of other corresponding F100AxM-0.025 as-prepared fiber specimens with the same contents of acid treated ATP nanofibers treating with HCl concentrations other than 5 M for 60 min.

Fig. 5
figure 5

DSC thermograms of a F100, b F100A0.0125, c F100A0.025, d F100A0.05, e F100A0.15, f F100A0.25 as-prepared fiber specimens

Full size image
Fig. 6
figure 6

DSC thermograms of a F100, b F100A1M-0.0125, c F100A1M-0.025, d F100A1M-0.05, e F100A1M-0.15, f F100A1M-0.25 as-prepared fiber specimens

Full size image
Fig. 7
figure 7

DSC thermograms of a F100, b F100A2M-0.0125, c F100A2M-0.025, d F100A2M-0.05, e F100A2M-0.15, f F100A2M-0.25as-prepared fiber specimens

Full size image
Fig. 8
figure 8

DSC thermograms of a F100, b F100A3M-0.0125, c F100A3M-0.025, d F100A3M-0.05, e F100A3M-0.15, f F100A3M-0.25 as-prepared fiber specimens

Full size image
Fig. 9
figure 9

DSC thermograms of a F100, b F100A5M-0.0125, c F100A5M-0.025, d F100A5M-0.05, e F100A5M-0.15, f F100A5M-0.25 as-prepared fiber specimens

Full size image
Fig. 10
figure 10

DSC thermograms of a F100, b F100A7M-0.0125, c F100A7M-0.025, d F100A7M-0.05, e F100A7M-0.15, f F100A7M-0.25 as-prepared fiber specimens

Full size image
Table 5 Melting temperatures and evaluated lamellar thickness values of UHMWPE as-prepared fiber specimens
Full size table

As evidenced by the specific surface area and SEM analyses, the purified and acid treated ATP nanofibers are with relatively large surface areas per volume, which make them in close proximity to a large fraction of the UHMWPE matrix. Apparently, even small contents of purified and acid treated ATP nanofibers can serve as efficient nucleation sites for UHMWPE molecules during their gel-spinning processes. These efficient nucleation sites of purified and acid treated ATP nanofibers then facilitate the crystallization of UHMWPE molecules into crystals with thinner lamellar thickness and/or lower Tm values during their gel-spinning processes. Acid treated ATP nanofibers with even higher specific surface areas are likely to serve as more effective sites for nucleation of UHMWPE molecules than purified ATP nanofibers. As a consequence, F100A5M-y as-prepared fiber specimens exhibit significantly higher Xc but lower Tm and/or lamellar thickness values than the corresponding F100Ay and/or other F100AxM-y as-prepared fiber specimens with the same purified and/or acid treated ATP contents, respectively.

Orientation factor

Typical orientation factor (fo) values of F100, F100Ay and F100AxM-y as-prepared and drawn fiber series specimens are summarized in Fig. 11. No significant difference in fo values was found for F100, F100Ay and F100AxM-y as-prepared fiber specimens. As expected, fo values of F100 and each F100Ay or F100AxM-y fiber series specimens increase consistently as their draw ratios increase. After addition of purified and/or acid treated ATP nanofibers, the fo values of drawn F100Ay and/or F100AxM-y fiber specimens are significantly higher than those of F100 fiber specimens with the same draw ratios. The fo values of drawn F100Ay and F100AxM-y fiber specimens with the same draw ratios reach a maximal value as their purified and acid treated ATP contents approach the optimal values at 0.05 and 0.025 phr, respectively. In which, fo values of drawn F100AxM-y fiber specimens are higher than those of the corresponding F100Ay fiber specimens with the same draw ratios and ATP contents but without treating with HCl solutions. Furthermore, it is worth to note that the fo values of drawn F100AxM-0.025 fiber specimens prepared at the optimal content of acid treated ATP nanofibers reach another maximal value as the HCl concentration and treating time approach the optimal value at 5 M and 60 min, respectively.

Fig. 11
figure 11

The orientation factor (fo) values of F100 (+), F100A0.0125 (), F100A0.025 (◇), F100A0.05 (△), F100A0.15 (×), F100A0.25 (○), F100A1M-0.0125 (□), F100A1M-0.025 (◇), F100A1M-0.05 (△), F100A1M-0.15 (×), F100A1M-0.25 (○), F100A2M-0.0125 (□), F100A2M-0.025 (◇), F100A2M-0.05 (△), F100A2M-0.15 (×), F100A2M-0.25 (○), F100A3M-0.0125 (□), F100A3M-0.025 (◇), F100A3M-0.05 (△), F100A3M-0.15 (×), F100A3M-0.25 (○), F100A5M-0.0125 (□), F100A5M-0.025 (◇), F100A5M-0.05 (△), F100A5M-0.15 (×), F100A5M-0.25 (○), F100A7M-0.0125 (□), F100A7M-0.025 (◇), F100A7M-0.05 (△), F100A7M-0.15 (×), F100A7M-0.25 (○) as-prepared and drawn fiber specimens with varying draw ratios

Full size image

Achievable draw ratios of the as-prepared fiber specimens

Figure 12 summarized the achievable draw ratios (Dra) of the F100, F100Ay and F100AxM-y as-prepared fibers prepared at varying purified and acid treated ATP contents. Similar to those found in our previous study [41], Dra values of F100Ay as-prepared fibers are significantly higher than that of the F100 as-prepared fiber specimen without addition of purified ATP nanofibers. The Dra values of F100Ay as-prepared fiber series specimens increase significantly with the initial increase in purified ATP contents and then reach a maximal value at 145 as their purified ATP contents approach an optimal value at 0.05 phr. After treatment of HCl solutions on purified ATP nanofibers, the Dra values of each F100AxM-y as-prepared fiber series specimens are significantly higher than those of the corresponding F100Ay as-prepared fiber specimens with the same purified ATP contents but without acid treatment. Similar to those found for F100Ay as-prepared fiber specimens, the Dra values of each F100AxM-y as-prepared fiber series specimens reach a maximal value as their acid treated ATP contents approach a lower optimal value at 0.025 phr. Moreover, it is worth noting that the maximal Dra values of the F100AxM-0.025 as-prepared fiber specimens reached another maximal value as their purified ATP nanofibers were treated by 5 M HCl solutions for 60 min. As shown in Fig. 12, the maximal Dra values obtained for F100A5M-0.025 as-prepared fiber specimen reaches 249, which is significantly higher than those of F100AxM-0.025 as-prepared fiber specimens prepared with 0.025 phr ATP nanofibers treated with HCl concentrations other than 5 M.

Fig. 12
figure 12

Achievable draw ratios of F100 (◆), F100Ay (□), F100A1M-y (△), F100A2M-y (×), F100A3M-y (◇), F100A5M-y (○) and F100S7M-y (+) as-prepared fiber specimens

Full size image

As evidenced by thermal and lamellar thickness analyses, the Tm and/or evaluated lc values of F100Ay and F100AxM-y as-prepared fiber specimens reduce significantly with the increase in purified and/or acid treated ATP contents, although the amounts of crystals with lower Tm and/or evaluated lc values increase significantly as their purified and/or acid treated ATP contents increase. Presumably, these crystals with lower Tm and/or evaluated lc values obtained at higher purified and/or acid treated ATP contents can be melted and pulled out of folded lamellar crystals relatively easily during ultradrawing processes, and hence, results in higher drawability and orientation of the F100Ay and F100AxM-y fibers. However, the amounts of coagulated and/or reticulated purified ATPs and/or acid treated ATPs are likely to increase significantly when their ATPs contents are higher than certain values. These coagulated and/or reticulated purified ATPs and/or acid treated ATPs can slide with each other and serve as the defects during the drawing processes of F100Ay and F100AxM-y as-prepared fiber specimens. The defects are likely to hinder the drawing and orientation of crystallites of the ultradrawn fibers along the drawing direction, and hence lead to a significant reduction in their achievable draw ratio and orientation factor values. Based on these premises, it is reasonable to understand that the achievable draw ratios of F100Ay and F100AxM-y as-prepared fiber specimens and orientation factor values of the F100Ay and F100AxM-y fibers with a fixed draw ratio reduce significantly when their purified and/or acid treated ATP contents are higher than the specific optimal values.

Morphological analyses of the as-prepared and drawn fibers

Figure 13 show typical SEM micrographs of the as-prepared and drawn F100, F100A0.025 or F100A5M-0.025 fiber specimens with various draw ratios. Many demarcated drawn “micro-fibrils” were found paralleling the drawing direction of the drawn F100, F100A0.025 or F100A5M-0.025 fiber specimens as their draw ratios increase, wherein the thicknesses of these drawn micro-fibrils reduce significantly as the draw ratios increase. Moreover, it is worth noting that more and thinner “micro-fibrils” were found on the etched surfaces of F100A5M-0.025 fiber specimen than those of F100A0.025 and F100 fiber specimens with the same draw ratios, respectively.

Fig. 13
figure 13

SEM micrographs of the F100 fiber specimens with a draw ratio of a 1, b 50, c 100; and of the F100A0.025 fiber specimens with a draw ratio of d 1, e 50, f 100; and of the F100A5M-0.025 fiber specimens with a draw ratio of g 1, h 50, i 100

Full size image

It is not completely clear what accounts for the interesting demarcated “micro-fibril” morphology found in the F100, F100A0.025 or F100A5M-0.025 drawn fiber specimens. Presumably, during the ultra-drawing processes, many of the UHMWPE kebab crystals with relatively thinner lamellar thickness values can be unfolded and pulled out of the crystal lamellae in an easier way than those kebab crystals with thicker lamellar thickness values. The unfolded UHMWPE molecules pulled out from the kebab lamellae can then gradually transform into the oriented “micro-fibrils” during their ultra-drawing processes. As evidenced by DSC analysis in the previous section, the Tm and/or evaluated lc values of F100Ay and F100AxM-y as-prepared fiber series specimens reduce significantly as their purified ATP and/or acid treated ATP contents increase. In which, crystals with rather low Tm and evaluated lc values were found for F100AxM-y fiber specimens than those of the corresponding F100Ay with the same ATP contents but without acid treatment. It is, therefore, reasonable to understand that the “micro-fibrils” found on the etched surfaces of drawn F100A5M-0.025 and F100A0.025 fiber specimens are more and thinner than those of the corresponding drawn F100 fiber specimens with the same draw ratios. In which, more and thinner “micro-fibrils” were found on the etched surfaces of F100A5M-0.025 fiber specimen than those of F100A0.025 and F100 fiber specimens with the same draw ratios.

Tensile properties

The tensile properties of F100, F100Ay and F100AxM-y series fiber series specimens prepared at varying draw ratios are illustrated in Figs. 14 and 15 (Table 6). As expected, the tensile strength (σf) and modulus (E) values of the drawn F100, F100Ay and F100AxM-y fiber series specimens improve consistently as their draw ratios increase. Similar to those found for their orientation factors, the σf and E values of both drawn F100Ay and F100AxM-y fiber series specimens with a fixed draw ratio reach a maximal value as their purified and/or acid treated ATP contents approach the optimal values at 0.05 and 0.025 phr, respectively, wherein the σf and E values of the F100AxM-y fiber specimens are always significantly higher than those of the corresponding F100Ay fiber specimens prepared at the same draw ratios and ATP contents but without acid treatment.

Table 6 Tensile strength (σf) and modulus values (E) of UHMWPE drawn fiber specimens
Full size table
Fig. 14
figure 14

The tensile strengh values of F100 (+), F100A0.0125 (□), F100A0.025 (◇), F100A0.05 (△), F100A0.15 (×), F100A0.25 (○), F100A1M-0.0125 (□), F100A1M-0.025 (◇), F100A1M-0.05 (△), F100A1M-0.15 (×), F100A1M-0.25 (○), F100A2M-0.0125 (□), F100A2M-0.025 (◇), F100A2M-0.05 (△), F100A2M-0.15 (×), F100A2M-0.25 (○), F100A3M-0.0125 (□), F100A3M-0.025 (◇), F100A3M-0.05 (△), F100A3M-0.15 (×), F100A3M-0.25 (○), F100A5M-0.0125 (□), F100A5M-0.025 (◇), F100A5M-0.05 (△), F100A5M-0.15 (×), F100A5M-0.25 (○), F100A7M-0.0125 (□), F100A7M-0.025 (◇), F100A7M-0.05 (△), F100A7M-0.15 (×), F100A7M-0.25 (○) fiber specimens with varying draw ratios

Full size image
Fig. 15
figure 15

The modulus values of F100 (+), F100A0.0125 (□), F100A0.025 (◇), F100A0.05 (△), F100A0.15 (×), F100A0.25 (○), F100A1M-0.0125 (□), F100A1M-0.025 (◇), F100A1M-0.05 (△), F100A1M-0.15 (×), F100A1M-0.25 (○), F100A2M-0.0125 (□), F100A2M-0.025 (◇), F100A2M-0.05 (△), F100A2M-0.15 (×), F100A2M-0.25 (○), F100A3M-0.0125 (□), F100A3M-0.025 (◇), F100A3M-0.05 (△), F100A3M-0.15 (×), F100A3M-0.25 (○), F100A5M-0.0125 (□), F100A5M-0.025 (◇), F100A5M-0.05 (△), F100A5M-0.15 (×), F100A5M-0.25 (○), F100A7M-0.0125 (□), F100A7M-0.025 (◇), F100A7M-0.05 (△), F100A7M-0.15 (×), F100A7M-0.25 (○) fiber specimens with varying draw ratios

Full size image

The mechanical properties of the drawn specimens are generally believed to depend mainly on the degree of orientation of the drawn specimens, as their molecular weights are constant [48, 49]. As evidenced by the orientation analysis, at a fixed draw ratio, the orientation factor values of the drawn F100A0.05 and F100AxM-0.025 fiber specimens prepared at the optimal purified and acid treated ATP contents are always higher than those of other F100Ay and/or F100AxM-y fiber specimens prepared at ATP and/or acid treated ATP contents deviating from their optimal values, respectively. In which, the orientation factor values of the F100AxM-0.025 fiber specimens are always higher than those of the corresponding F100A0.05 fiber specimens with the same draw ratios and ATP contents but without acid treatment. These results clearly suggest that a good orientation of UHMWPE molecules along the drawing direction positively affects the tensile properties of the F100Ay and F100AxM-y fibers. Excellent orientation and tensile properties of the UHMWPE fibers can be prepared by ultradrawing the F100A0.05 and F100AxM-0.025 as-prepared fibers prepared at the optimal purified and/or acid treated ATP contents, respectively, in which, significant better orientation and tensile properties can be obtained for F100AxM-0.025 fiber specimens with even higher specific surface areas of acid treated ATP nanafibers.

Conclusions

The rodlike feature of acid treated ATP nanofibers were significantly etched into broken/shorter and thinner nanofibers by HCl solutions as their concentrations increase. In fact, after etching by 7 M HCl solution for 60 min, nearly only broken/short ATP nanofibers were found for A7M specimen. Similarly, A5M specimens were etched into broken/shorter and thinner rodlike feature as the treating time increased from 20 to 100 min. In which, nearly only broken and short ATP nanofibers were found for A5M specimens after they were excessive etched for more than 60 min. The specific surface area of the purified ATP specimen is relatively high at 205 m2/g. The specific surface areas of each acid treated ATP series specimens reach a maximal value as the treating time approaches an appropriate value at 60 min. The maximal specific surface areas obtained for acid treated ATP specimens reach another maximal value at 282 m2/g as the HCl concentrations used for 60 min of acid treatment approach an optimal value at 5 M. The Tm(or evaluated lc) and Xc values of F100Ay and/or F100AxM-y as-prepared fiber series specimens approach the minimal and maximal values, as their purified and/or acid treated ATP approach the optimal values at 0.05 and 0.025 phr, respectively. In which, Tm(or evaluated lc) and Xc values of the F100AxM-y as-prepared fiber specimens are significantly lower and higher than those of the corresponding F100Ay as-prepared fiber specimens with the same ATP contents but without acid treatment, respectively. Moreover, it is worth noting that each F100A5M-y as-prepared fiber series specimen exhibits the lowest Tm(or evaluated lc) but the highest Xc values among F100AxM-y as-prepared fiber series specimens with the same contents of acid treated ATP nanofibers treating with different HCl concentrations for 60 min. Apparently, the efficient nucleation sites of purified and acid treated ATP nanofibers facilitate the crystallization of UHMWPE molecules into crystals with lower Tm (or evaluated lc) values during their gel-spinning processes. Acid treated ATP nanofibers with higher specific surface areas appear to serve as more effective sites for nucleation of UHMWPE molecules than purified ATP nanofibers.

In a way similar to those found for the orientation factor values, the achievable draw ratios of the F100Ay and F100AxM-y as-prepared fibers and tensile strength values of their drawn fibers approach a maximal value as their purified ATP and/or acid treated ATP contents reach the optimal values at 0.05 and 0.025 wt%, respectively. The maximal Dra values of F100AxM-0.025 as-prepared series fiber specimens and the tensile strengths and moduli of the drawn F100AxM-0.025 fiber specimens are significantly higher than those of the F100A0.05 drawn fiber specimens prepared at the same draw ratios and optimal ATP contents but without acid treatment. Moreover, it is worth noting that the maximal Dra values of the F100AxM-0.025 as-prepared fiber specimens and the tensile strengths and moduli of the drawn F100AxM-0.025 fiber specimens reached another maximal value as their ATP nanofibers were treated by 5 M HCl solutions for 60 min. Morphological analyses revealed that more and thinner “micro-fibrils” were present on the etched surfaces of F100A5M-0.025 fiber specimen than those of F100A0.025 and F100 fiber specimens with the same draw ratios, respectively. The above results suggest that excellent orientation and ultimate tensile properties of UHMWPE fibers can be obtained by ultradrawing the F100AxM-0.025 as-prepared fibers prepared at the optimal contents of acid treated ATP nanofibers treated with proper condition.