Performance Mapping toward Optimal Addition Levels and Processing Conditions for Different Types of Hydrocarbon Waxes Typically Used as External Lubricants in Sn-Stabilized PVC Pipe Formulations

Abstract

The mechanism of performance of hydrocarbon waxes used as external lubricants in typical PVC pipe formulations is not completely understood or defined. This study will benefit pipe manufacturers to tailor PVC processing conditions and wax addition levels to enhance the value of waxes with different molecular architectures in Sn-stabilized PVC formulations. A selection of waxes was comprehensively analyzed to understand how their structural and morphological differences impact the fusion behavior of a Sn-stabilized PVC compound. Results showed a high probability of fusion for waxes with melting points between 65 and 80 °C and a kinematic viscosity (at 135 °C) of less than 5 cSt across all experimental temperatures and concentration levels tested. The general probability of fusion increased with increasing temperature and decreasing wax concentration. Wax compositional differences only had a large impact on PVC fusion at lower experimental temperatures and high wax concentrations. The stable time and onset of degradation were found to be sensitive to experimental temperature but less so to wax concentration. High melting waxes had the lowest fusion torque but a narrow operating window in terms of concentration and temperature for a successful fusion to occur. Overall, fully refined paraffin waxes had the widest operating window and were less sensitive to a change in the wax concentration and experimental temperature than the rest of the waxes that were tested.

Introduction

Approximately 60 million metric tons of PVC polymers are produced every year making them the third most significant synthetic plastic after polyethylene and polypropylene in terms of volume.¹⁻⁴ Processed PVC plastic products are produced in two grades: rigid unplasticized PVC (uPVC) and flexible plasticized PVC. uPVC represents the largest market segment of the two grades and mainly finds application in construction of pipes and profiles such as doors and window frames.⁵ Flexible PVC incorporates the addition of plasticizers such as phthalates for use in plumbing, electrical cables, wiring, films, and sheets.⁶ In the United States, two-thirds of potable pipes are made from PVC.⁷ Ingredients that are used for the production of potable pipes are regulated and have to be certified by the National Sanitary Foundation (NSF).⁸ The NSF is a public health and safety organization that tests and certifies ingredients of products in contact with drinking water (NSF/ANSI 61).⁸ In addition, qualified ingredients and acceptable concentration limits allowed for use in production of potable pipes that are approved and are published by the NSF and in the Plastics Pipe Institute TR-2/PPI PVC Range composition documentation.⁹ Part A of TR-2 lists pre-qualified ingredients and PVC compounds proven to meet cell class 12454 according to ASTM D1784.

PVC has unique hardness and durability properties making it a desirable polymer in the construction industry.^1,10 Durability is an important characteristic for potable pipes as it needs to remain underground and function for more than 50 years. Furthermore, PVC products have good thermal stability and windows made from these versatile polymers are highly energy-efficient, have good chemical stability, are waterproof, and are resistant to environmental stress cracking. PVC products also have good electrical insulation properties, are recyclable, and have excellent organoleptic properties. PVC production is economically attractive, and all these properties are the main drivers for PVC demand growth, especially in construction use, where growing urbanization drives increasing infrastructure investment, increasing automotive production, and increasing consumable volumes of packaging and footwear.

Various additives are mixed with PVC during processing due to the high melt viscosity and low heat stability of neat PVC. These additives all serve specific functions during PVC extrusion and contribute to critical processing parameters of fusion time and torque, stable time, and final product properties such as surface characteristics and impact strength.¹⁰ The rheological and fusion behavior of PVC is governed by the complex morphological structure of the PVC resin, compound composition, and processing conditions. Heating PVC particles does not result in a homogeneous melt, rather powder particles fuse together in a multi-stage process. Fusion requires breaking the particulate structure with typically long processing times and high temperatures that introduce the risk of thermal degradation. Additives such as heat stabilizers and processing aids are introduced to prevent thermal degradation and improve melt strength, respectively. Fillers, UV stabilizers, and impact modifiers are also added depending on the final applications and required properties of the PVC extrudate.

Lubricants are considered to be one of the most important processing additives and parameters in PVC.¹¹ Lubricants are generally classified as external or internal depending on their affinity for the PVC matrix. External lubricants have less chemical association with the polymer matrix and migrate to the metal surface where they act as metal release agents, thereby delaying fusion. Internal lubricants have higher chemical affinity for the polymer matrix, promote inter-particle friction, and hence decrease fusion time. The internal and external lubricants generally have some degree of affinity for each other as well. Barnard et al. found that the use of an internal lubricant promotes the dispersion of the external lubricant in PVC.¹¹ Sufficient fusion is obtained through the optimal combination of internal and external lubricants, which is important for lowering operating pressure and torque, and consequently vital for the final impact strength and gloss of the PVC product. In recent years, various studies have been conducted to evaluate the performance of additives and understand the mechanism of fusion. These experiments included differential scanning calorimetry to evaluate the degree of fusion, torque rheometry, mechanical analyses to correlate fusion with material strength, scanning electron microscopy to analyze visual changes upon fusion, and an extruder screw freezing method to track the development of fusion during the extrusion process.¹¹⁻¹⁷ However, none of the recent reports comprehensively studied the effect of chemical and morphological differences of lubricants on the PVC fusion behavior. During this study, the impact of external lubricants on PVC fusion is evaluated.

The focus of this article is on external lubricants as described in Part A3 in TR-2 (Table 1) as hydrocarbon waxes and defined in this study as fully refined paraffin wax (FRP). The FRP waxes were compared to Part B2 functional equivalent waxes that do not meet Part A3 requirements and defined here as Fischer–Tropsch wax (FT) and linear alpha-olefin wax (LAO wax).⁹ Optimal PVC lubrication conditions associated with differences in wax architecture and addition levels are described. This study aims to benefit pipe manufacturers to tailor PVC processing conditions and addition levels to optimize the value of waxes with different molecular structures and morphologies.

Table 1. Properties of Hydrocarbon Waxes as Defined by TR-2⁹.

property	units	test method	requirements
chemical type		see footnotea	hydrocarbon waxa
congealing point	°F (°C)	ASTM D938	149–169 (65–76)
viscosity at 210 °F	cSt	ASTM D445	5.5–7.5
kinematic carbon number distribution of normal hydrocarbons	%	ASTM D5442	min 80% C26–C50
			<20% C26 and lower
			<10% above C50
			zero above C85
non-normal paraffin	%	ASTM D5442	10–50
needle penetration at 77 °F		ASTM D1321	10–18
oil content	%	ASTM D721	max 1%
flash point	°F (°C)	ASTM D92	449 (230)
color (Saybolt)		ASTM D156	min +10
acid number		ASTM D1386	max 0.5
density	g/cm3	ASTM D792	0.915–0.940
physical appearance			small uniform flake, prill or powderb

^aHydrocarbon waxes containing linear and branched chains with carbon numbers from C20 to C60.

^bThis requirement is not applicable if the wax is added as liquid.

Experimental Section

Materials

Suspension polymerized polyvinyl chloride (PVC, SE-950, Shintech Inc.) powder with a K-value of 65 was used. Calcium stearate (CaSt, Doverlube CA-21, Dover Chemical Corporation) was used as an internal lubricant. A tin-based heat stabilizer (Mark 1925, Galata Chemicals) was added to the formulation to prevent thermal degradation during mixing. Titanium dioxide was also included in the blends (HiTox TiO₂, TOR Minerals). The formulation details are summarized in Table 2. Thirteen waxes were included in the study (see the Supporting Information). The waxes represented the following types: fully refined paraffin wax (FRP), linear alpha-olefin wax (LAO), and Fischer–Tropsch wax (FT). Two proprietary wax blends based on these waxes were also included: Blend 1 is a blend of FRP and a high melt FT wax (FT-FRP), and Blend 2 is LAO wax blended with the same high melt FT wax (FT-LAO). A selection of these waxes and blends was chosen to represent the widest range of wax architectures. Table 3 lists these representative waxes included in this report and summarizes viscosity, melting point, and carbon number data.

Table 2. PVC Testing Formulation.

PVC (phr)	wax (phr)	calcium stearate (phr)	methyltin mercaptide(phr)	titanium dioxide(phr)
100	0.9 and 1.1	0.6	0.6	0.5

Table 3. Wax Properties.

wax	kinematic viscosity (cSt)a	drop melt point (°C)b	carbon number rangec	average carbon numberd
FRP1	3.1	65	C18–C46	C29
LAO	4.3	75	C28–C58	C30+e
FRP2	4.5	75	C20–C54	C35
FT1	5.0	89	C16–C75	C44
FT-FRP (Blend 1)	7.1	105	C22–C90	C56
FT-LAO (Blend 2)	9.6	87	C22–C90	C26+e; C56
FT2	16.1	116	C25–C90	C58

^aASTM D445 standard test method at 135 °C.

^bASTM D3954 standard test method.

^cCalculated from HT-GC.

^dAlso calculated from HT-GC.

^eLAO is a C30+ LAO wax, and FT-LAO is a blend of a C26+ LAO wax and a high melting FT wax with an average carbon number of C56.

Experimental Methods

High-Temperature Gas Chromatography (HT-GC)

The carbon distribution and the degree of branching were determined on a Hewlett-Packard 5890 gas chromatograph fitted with a split-splitless inlet system with constant pressure control, operated at 75 kPa throughout the analysis. A flame ionization detector (FID) was used for detection. The injector and detector were both operated at 400 °C. The instrument was fitted with a Restek MXT-1 capillary column (15 m × 0.28 mm, 0.1 μm), and helium was used as carrier gas. The programmable oven was programmed from a starting temperature of 60 °C, held for 2 min, raised by 10 °C/min to a final temperature of 430 °C, and held for 2 min. The samples were dissolved in cyclohexane before injection of 1 μL in splitless mode. For hard waxes, the oven was programmed from a starting temperature of 60 °C, held for 2 min, raised by 10 °C/min to a final temperature of 430 °C, and held for 15 min to separate and elute the heavy components.

Fourier Transform Infrared Spectroscopy (FTIR)

FTIR measurements were carried out on a Thermo Scientific Nicolet iS10 spectrometer (Waltham, USA) operating in attenuated total reflectance (ATR) mode with a spectral resolution of 4 cm^–1. The system was equipped with a diamond crystal, and 32 scans were recorded per sample between 600 and 4000 cm^–1 wavenumbers. A background spectrum was collected (32 scans) before each sample. Thermo Scientific OMNIC (version 9) software was used for processing of the collected data.

Nuclear Magnetic Resonance Spectroscopy (NMR)

Chemical compositions of the waxes were studied by solution ¹³C and ¹H NMR spectroscopies. Analyses were done on a 600 Varian Unity Inova NMR spectrometer equipped with an Oxford magnet operating at 600 MHz having a 5 mm inverse detection pulsed field gradient probe. Spectra were recorded at 120 °C, and samples were prepared by dissolving 60 mg of wax in deuterated 1,1,2,2-tetrachloroethane (TCE-d₂) in a NMR tube. TCE-d₂ was used as the internal reference at 74.5 ppm.

For clarity, the peaks in the ¹³C NMR spectra have been labeled only using the representative symbols, as denoted in Figure 1. Peak assignments for NMR follow the general nomenclature, where saturated end groups were denoted as “s” and unsaturated end groups were denoted as “u” with the corresponding number of the carbon increasing in the inward direction. In cases where branches occurred, the carbons on the branch were labeled numerically as would be done for end groups. Adjacent carbons are labeled using Greek notation in both directions of the olefin or branching carbon, except where one direction terminates into an end group within four carbons of the olefin or branching carbon.

Figure 1.

General structure elucidating the carbon naming conventions for solution NMR.

Solid-state NMR spectra were acquired using a Varian VNMRS 500 MHz two-channel spectrometer using 6 mm zirconia rotors and a 6 mm Chemagnetics T3 HX MAS probe. The experimental parameters were optimized using a KBR/Adamantane standard. Proton (¹H) wideline experiments were performed at ambient temperature in static mode. All samples were analyzed in a similar manner for comparison purposes.

Differential Scanning Calorimetry (DSC)

Endothermic melting and exothermic crystallization behavior were studied using DSC under an inert nitrogen (N₂) atmosphere and operating at a gas flow rate of 50 mL/min. Analyses were carried out using a TA Instruments Q100 calorimeter, which was calibrated with an indium metal standard according to standard procedures. Samples were subjected to three thermal cycles. In the first heating cycle, the specimens were heated up to 150 °C to erase the thermal history. The second cycle consisted of a non-isothermal cooling step from 150 down to −10 °C. During the third cycle, samples were reheated to 150 °C. The second heating cycles were used to calculate melting temperatures and enthalpies. Heating and cooling rates were fixed at 10 °C/min.

Scanning Electron Microscopy (SEM)

The samples were analyzed using a Zeiss EVO scanning electron microscope. Prior to imaging, the samples were mounted on aluminum stubs with double-sided carbon tape. The samples were coated with a thin (∼10 nm thick) layer of gold using an Edwards S150A gold sputter coater. A Zeiss five-diode backscattered electron (BSE) detector (Zeiss NTS BSD) and Zeiss Smart SEM software were used to generate BSE images. The samples were chemically quantified by semiquantitative energy-dispersive X-ray spectrometry (EDS) using an Oxford Instruments X-Max 20 mm² detector and Oxford Aztec software. Beam conditions during the quantitative analysis and backscattered electron image analysis on the Zeiss EVO were 20 kV accelerating voltage, 8 nA probe current, with a working distance of 8.5 mm, and a beam current of 5 nA. The counting time was 10 s live-time.

Preparation of PVC Blends

PVC resin and a heat stabilizer were dry-mixed at room temperature at high speed for 15 min using an overhead paddle stirrer. Thereafter, the wax, calcium stearate, and titanium dioxide were added and mixed for an additional 15 min to ensure that all additives were homogeneously dispersed in the blend.

Fusion and Stable Time Determination

Pre-weighed (65 g) PVC blends were transferred to a pre-heated mixing chamber of a Thermo Scientific Haake Polylab OS torque rheometer coupled to a Haake Rheomix OS lab mixer and fitted with roller rotors. Analyses were conducted until a distinct onset of degradation was observed at which point the screw rotation was stopped, the sample was removed, and the mixing chamber was properly cleaned before the following sample was introduced. Details regarding formulation recipes and experimental temperatures are captured in the Results and Discussion section of this report. Sample mass and rotational rotor speed were fixed at 65 g and 65 rpm, respectively. Fusion time is the peak maximum after the powder peak on the resulting fusion curve, and stable time is the time between the fusion peak and the onset of degradation.

Results and Discussion

Wax Characterization

The eight wax samples were evaluated to understand their chemical composition and structural morphology. The aim was to ascertain whether differences in structure, composition, and morphology have significant effects on PVC fusion.

Figure 2 shows a summary of the FTIR and NMR results for all the samples evaluated in the study. The FTIR spectra for the fully refined paraffin waxes as well as the FT group of waxes contained bands characteristic of predominantly linear hydrocarbons supported by ¹³C NMR spectra that mainly showed the methylene backbone and saturated end group peaks. On closer inspection, the fully refined paraffin wax samples (FRP1 and FRP2) showed evidence of low level, isolated methyl branching, as shown in the representative FRP spectrum in Figure 3. Furthermore, the lower intensity of the end group peaks on the ¹³C NMR spectrum for FT2 indicates that it had a significantly higher molecular weight relative to all the other waxes analyzed. The double-bond peaks apparent on the FTIR spectrum of the LAO wax were quantified with NMR to be approximately 90% alpha olefins and 10% internal and branched olefins.

Figure 2.

(a) FTIR and (b) ¹³C solution NMR spectra of the waxes.

Figure 3.

Expanded ¹³C solution NMR spectrum illustrating the low-level methyl branching of the FRP waxes.

The FTIR spectra in Figure 2 also display the splitting of the peak at around 720 cm^–1 into a doublet, indicating the presence of crystalline order. With increasing disorder, the peak broadens and the splitting becomes less distinct. FT1 has the most pronounced intensity and splitting indicating significant long-range order in this sample due to its high chain linearity and homogeneity as identified with solution NMR. The solid-state proton wideline experiment yielded insights into the motional dynamics of the waxes. Linear hydrocarbon waxes are typically very rigid, and this rigidity can directly be associated with wax crystallinity, as already inferred from FTIR results. Figure 4 shows a comparison of the wideline spectra for all the waxes. The broadened line shape at the base of the peaks is indicative of long-range order associated with large crystallites. This is apparent on all spectra associated with FT as well as the FT-FRP blend. FRP waxes showed a characteristic broadening of the line shape, but the profile implies less distinct differences between the rigid and amorphous fractions in the sample on the timescale of the experiment and is therefore indicative of thin, interwoven, and more mobile crystallites. LAO exhibited an even narrower linewidth wideline spectrum. The presence of double bonds and especially internal olefins in the LAO wax disrupts crystal growth resulting in small, highly mobile crystallites. The FT-LAO blend showed morphological characteristics of both FT and LAO waxes.

Figure 4.

¹H wideline NMR spectra of the waxes.

Figure 5 shows a stacked plot of the melting endotherms of the waxes. FRP1 and FT1 had a single, sharp thermal transition due to their narrow, more homogeneous chemical composition distributions. The multiple melting peaks observed for the other waxes are indicative of compositional heterogeneity (high branch content for FRP2 and broad carbon number distribution for FT2) that lead to multiple crystallite sizes and a more complex morphology. In general, the melting temperatures follow closely the average carbon numbers and therefore the molecular weights of the waxes.

Figure 5.

DSC melting endotherms of (a) waxes with viscosities below 5 cSt and (b) waxes with viscosities above 5 cSt.

Wax Dispersion in the PVC/Wax/CaSt Mixture

To specifically assess the dispersion of the wax lubricant in the PVC formulation, PVC-lubricant blends were prepared at higher-than-usual concentration levels to enhance elemental signals for easier mapping. Concentrations of the PVC:wax:CaSt:stabilizer were kept constant at 100:5:3:3 phr, respectively. Blends were mixed at temperatures above the melting point of the waxes to facilitate mobility of the waxes as well as “emulating” the mixing zone in the extruder during processing. After 15 min of mixing, samples were allowed to cool to room temperature and prepared for SEM-EDS microscopy. Elemental mapping results are shown in Figure 6. Kinematic viscosities of the neat waxes are shown next to the wax labels. Images illustrate the dispersion of the respective waxes across the surface of the PVC particles after a high-speed melt mixing process, as discussed in the Experimental Section of this report. Sites where the lubricants were located are indicated by red signals on the electron images. Red signals indicate a high concentration of carbon (C) elements due to the hydrocarbon wax structures and allowed for visual determination of wax dispersion.

Figure 6.

SEM-EDS images of PVC formulations.

The FRP, LAO, and FT1 neat waxes showed homogeneous distributions of wax across the PVC particles, indicating that the calcium stearate facilitated the associations of external lubricants and the PVC. These observations are in good agreement with findings reported by Barnard et al. where less homogeneous apolar wax dispersions were observed in the absence of calcium stearate.¹¹ The dispersion homogeneity decreased for the higher viscosity wax (FT2), possibly due to the lower association of the longer chains with the aliphatic moiety of the internal calcium stearate lubricant. The most heterogeneous distribution was seen for the FT2 formulation.

The Impact of Wax on PVC Fusion Time, Torque, and Stability

PVC blends representing a typical resin:wax:CaSt:stabilizer ratio were formulated with the various waxes, as set out in the Experimental Section, to evaluate their effects on formulation fusion behavior and formulation stability. Comparative fusion studies were conducted at various concentrations and temperatures. Two wax concentrations (0.9 and 1.1 phr) and three different experimental temperatures (180, 185, and 190 °C) consistent with typical commercial operating conditions are discussed here. Table 4 shows a summarized dataset of all the fusion experiments performed under aforementioned conditions reporting fusion time, torque, and stable time, and Figure 7 shows the fusion curves of all successful experiments. FRP1 was not measured at 190 °C since very short fusion times of less than 2 min were expected in line with the trends observed. PVC manufacturing practices typically consider fusion times between 2 and 5 min as optimal; therefore, since fusion was observed at less than 2 min at 185 °C, it was not necessary to study this wax further at 190 °C.

Table 4. Summary of Fusion Time and Fusion Torque under Various Conditions^a.

wax	viscosity (cSt)		180 °C	185 °C	190 °C		180 °C	185 °C	190 °C
Fusion time (min)
FRP1	3.1	0.9 phr	4.0	1.4	DNM	1.1 phr	3.0	2.2	DNM
LAO	4.3		2.9	1.4	1.0		6.0	3.5	1.8
FRP2	4.5		5.4	3.0	2.0		4.8	2.5	1.6
FT1	5.0		NF	3.7	1.8		NF	NF	NF
FT2	16.1		NF	NF	NF		NF	NF	NF
Fusion torque (Nm)
FRP1	3.1	0.9 phr	24	28	DNM	1.1 phr	23	24	DNM
LAO	4.3		25	28	30		20	19	21
FRP2	4.5		20	21	22		21	20	21
FT1	5.0		NF	19	22		NF	NF	NF
FT2	16.1		NF	NF	NF		NF	NF	NF
Stable time (min)
FRP1	3.1	0.9 phr	5.1	5.3	DNM	1.1 phr	5.8	5.2	DNM
LAO	4.3		5.4	5.0	4.3		5.3	5.0	4.3
FRP2	4.5		5.5	5.0	4.6		7.9	5.4	4.5
FT1	5.0		NF	4.0	3.5		NF	NF	NF
FT2	16.1		NF	NF	NF		NF	NF	NF

^aNF = no fusion observed, DNM = did not measure under the specific conditions.

Figure 7.

Summary of fusion curves for all successful experiments.

Fusion times were affected by experimental temperatures and wax concentration. A decrease in fusion time was observed as the experimental temperature increased. Higher temperatures resulted in higher thermal energy that facilitates the migration of the lubricant, thereby promoting interparticle fusion, resulting in shorter fusion times. The standard deviation of the fusion time also changes from ±1 min at the lowest temperature to ≈0.5 min at the highest temperature for the same wax. An increased wax concentration typically delayed fusion time due to a higher degree of lubrication and decreased inter-particle friction in the blend during mixing. More reproducible fusion times were visible at high temperatures for different waxes, inferring that underlying chemical and morphological differences of these waxes play out less under these conditions (indicated by the arrows in Figure 7). On the other hand, larger differences in fusion times were observed at lower experimental temperatures, more so at high wax loadings, clearly indicating the differentiated contribution of the wax characteristics toward the fusion behavior. Melt viscosities of the waxes appeared to be a driving factor during PVC fusion, where waxes with viscosities below 5 cSt (at 135 °C) resulted in the most versatile formulations and promoted fusion under widespread conditions. Within this group, the LAO was more sensitive to changes in temperatures and concentration than the two FRP-based formulations. LAO had the shortest fusion time at 0.9 phr and 180 °C and the longest fusion time at 1.1 phr and 180 °C. The olefin nature of the wax could influence this unpredictable behavior, which may cause complications during commercial production. The kinematic viscosity of FT1 is at a threshold of 5 cSt, which means that, under favorable conditions, its fusion behavior is similar to that of generic-type waxes but more sensitive to changes in temperature and concentration than generic waxes.

It is clear that the higher viscosity FT resulted in delayed fusion. This is very evident not only in the formulations where FT2 is used but also in FT1 that has only slightly higher viscosity than the other waxes in the group. At 1.1 phr, no fusion was observed for higher viscosity FT waxes at temperatures ≤190 °C, suggesting that a much higher thermal energy is needed to promote fusion when using these lubricants. The trend for the high viscosity waxes is further substantiated in the fusion summary (bubble plots) later in the manuscript.

Figure 8 shows the summarized overlays of stable time for all formulations, under similar conditions discussed in Figure 7. In this study, the stable time was defined as the elapsed time between the fusion peak maximum and the onset of degradation. No direct correlation between stable time and fusion time was observed. The lower viscosity wax formulations (KV ≤ 5 cSt) had stable times within 1 min of each other throughout the experimental temperatures tested despite differences in fusion times. The higher viscosity FT formulations showed distinctly lower stable times due to the higher experimental temperatures required for fusion. Furthermore, the stable times were more dependent on experimental temperatures and less dependent on the wax concentration. This is evident by the decrease when moving from left to right (increase in temperature; 180–190 °C) in Figure 8 compared to the negligible difference when moving from top to bottom (increase in the wax concentration; 0.9–1.1 phr). The decreased stable times can be attributed to the earlier onset of degradation at elevated temperatures.

Figure 8.

Summary of stable times for all successful experiments.

The Impact of Wax Blends on PVC Fusion

Table 3 shows that FT-FRP has a lower viscosity but higher drop melt point than FT-LAO. This is likely because the melting point for FT-FRP is dominated by the higher molecular weight components of the blend, while the LAO component in FT-LAO is causing it to start melting earlier. Figures 9–12 correlate fusion behavior and molecular characteristics of the wax blends. Figure 9 shows the fusion curve overlays of PVC blends formulated with the two wax blends as well as PVC blends formulated with the individual wax components (FT, FRP, and LAO) at a wax concentration of 1 phr and experimental temperature of 187.5 °Ca. Both neat LAO and neat FRP resulted in fusion below 3 min (dotted black curves). Fusion occurred more rapidly for the LAO formulation and also followed the kinematic viscosity trend of the waxes as discussed in the previous section. In contrast, no fusion was obtained when only the higher viscosity and higher melting FT wax were incorporated (solid black curves). Interestingly, the FT-FRP wax blend promoted delayed fusion under these conditions whereas the FT-LAO blend behaved much like the neat FT formulation. The influence of the FT component could be seen by the delayed fusion and flatter fusion plateau region (indicative of higher melt viscosities) for FT-FRP than the neat FRP profile. Other concentrations were also evaluated, but no fusion was observed at 180 °C, evident of the influence of the higher melting FT wax component present in both wax blends. Fusion was obtained for both blends at 0.9 phr and temperatures of 185 and 190 °C.

Figure 9.

Fusion curve overlays of blends with their individual blend components for (a) FT-FRP and (b) FT-LAO blends.

Figure 12.

SEM-EDS images of PVC blends formulated with FT-FRP, FT-LAO, and FT2 waxes.

Chromatographic analysis by HT-GC revealed a larger overlap in carbon number distribution between the individual FT and FRP components in Blend 1 than the individual FT and LAO components in Blend 2. Figure 10 highlights the overlapping distributions. The minimal overlap in Blend 2 could result in lower compatibility between individual wax components and hence the dominant FT effect seen in the fusion behavior. SEC chromatograms and DSC thermograms in Figure 11 substantiate the GC data. A larger overlap in molecular weight distributions (MWD) was observed for FT and FRP, and Blend 1 showed a unimodal MWD. A distinct shoulder on the higher molecular weight end was seen for Blend 2. The merging of endothermic DSC melting peaks for Blend 1 as seen in Figure 11c suggests better co-crystallization within the FT-FRP blend. Less observable DSC shifts were seen for FT-LAO, characteristic of less compatible blend components.

Figure 10.

HT-GC chromatograms of (a) FT-FRP and (b) FT-LAO blend components.

Figure 11.

HT-SEC (a, b) and DSC (c, d) overlays of blends with their individual blend components.

The evidence suggests that better compatibility between components of FT-FRP allowed this blend to fuse under conditions where FT and FT-LAO did not promote fusion. The FRP component essentially acted as a performance enhancer, enabling fusion, as the neat FT did not promote fusion under similar conditions. The influence of the higher melting FT component still played out in delayed fusion and increased blend viscosity, clearly seen from the slope of the fusion curve during the mixing zone. It is therefore crucial to understand the role of the individual components when formulating wax blends to optimize fusion performance. The FT-FRP blend has significantly lower viscosity and, consequently, slightly better dispersion than the FT-LAO blend. This could be explained by the improved compatibility of FT-FRP compared to FT-LAO. Both blends showed better dispersion relative to the neat high melting FT system, as seen in Figure 12.

The Impact of Calcium Stearate on PVC Fusion

Fusion behavior of binary PVC and calcium stearate (CaSt) blends, in the absence of wax, is compared to a PVC blend containing FRP2 wax in Figure 13. Under all conditions, fusion for the binary PVC/CaSt blend occurred rapidly and below 1.5 min. This shows the significant role that the wax lubricant plays during PVC extrusion and to make the process more manageable by delaying the onset of fusion. In addition, the PVC/CaSt blends resulted in significantly higher fusion torques than the blend containing wax as the calcium stearate only results in limited slip between PVC particles. This further highlights the importance of the wax in reducing melt viscosity of the formulations, reducing friction during extrusion and hence lowering the torque. The reduction in torque is commercially advantageous for the preservation and longevity of instrumentation. Figure 13 further highlights the reproducibility of the FRP formulation as discussed in the preceding sections.

Figure 13.

Fusion curves of a PVC/CaSt binary blend compared to a formulation containing FRP2 wax.

Summary of all Fusion Experiments

The bubble plots in Figure 14 summarize the outcomes for fusion times and fusion torques of 64 successful fusion experiments. The wax concentration and experimental temperature were varied between 0.7 and 1.3 phr and 175 and 200 °C, respectively. The bubble colors represent the three wax groups, namely, Group 1 (black, the low-level branched fully refined paraffin waxes and LAO waxes); Group 2 (red, the linear FT waxes); and Group 3 (green, wax blends of Groups 1 and 2, e.g., Blends 1 and 2). The size of the bubble represents the wax concentration. This extensive dataset validates the findings from the preceding discussions. Group 1 waxes showed successful fusion over a broad concentration and temperature band and was therefore more versatile and robust toward operational changes. The high viscosity FT formulations only showed successful fusion at high temperatures and/or low concentrations and had significantly longer fusion times but generally more favorable torque values than the low viscosity waxes. Group 3 shows the benefit of blending different waxes, depending on wax selections and blend ratios, and formulations that could be tuned to be as versatile as Group 1 with the added benefit of lowering the fusion torque like Group 2 with improved stable times.

Figure 14.

Bubble plots of 64 successful fusion experiments summarizing (a) fusion time and (b) fusion torque.

Conclusions

The performance of different hydrocarbon waxes used as external lubricants in typical PVC pipe formulations was successfully investigated, and a better understanding of the impact of the lubricant on fusion behavior was obtained. The results showed a high probability of fusion across all experimental temperatures and concentration levels tested for Group 1 waxes, which typically had melting points between 65 and 80 °C and a kinematic viscosity (at 135 °C) of less than 5 cSt.

The general probability of fusion increases with increasing temperature and decreasing wax concentration for all types of waxes tested. Above 190 °C fusion occurred, irrespective of the viscosity and molecular differences of the waxes. Wax compositional differences had a large impact on PVC fusion only at lower experimental temperatures and high wax concentrations. The study highlighted the importance of compatibility of blend components. When blends consist of significantly different molecular structures, it leads to unpredictable fusion outcomes and less predictive performance and properties.

For all successful fusion results, the fusion torque for higher melting, high viscosity waxes were generally lower than the rest but concentration-dependent. The fusion torque was higher for low melting waxes but insensitive to a change in concentration or temperature. The stable time and onset of degradation were found to be more sensitive to experimental temperature than the wax concentration. A decrease in stable time accompanied by an earlier onset of degradation was observed for all waxes with increasing temperature.

High melting waxes had the lowest torque but generally required low concentrations and/or high temperatures for fusion to occur. The results showed that fully refined paraffin waxes have the widest operating window in terms of temperature and wax concentration and can be easily tuned to achieve the desired fusion response for optimal performance and properties.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c00392.

• Additional chemical composition and viscosity data of the waxes (PDF)

The authors declare no competing financial interest.