Impact of the Chain Extension of Poly(Ethylene Terephthalate) With 1,3-Phenylene-bis-Oxazoline and N,N-Carbonylbiscaprolactam by Reactive Extrusion on its Properties.
Poly(ethylene terephthalate) (PET) is considered to be one of the most important technical polymers and the most important polyester due to the rapid growth of its usage in the past decades. The PET market increases currently very rapidly. In 2015, the market share of PET rose by 8% in Europe. It is estimated that the worldwide market for PET will amount to 22,726 kt, and 3,432 kt in Europe in 2017 . PET is mainly used for fibers and food packages. Its excellent properties like high tensile and impact strength, clarity, good processability, high chemical resistance, high thermal stability, and recyclability are the main reasons for the preferred usage of PET. Especially, the interest in recycling of PET increases continuously because of its high consumption. In Germany, the interest in recycling of postconsumer PET is very high due to its deposit system of one-way-bottles which has been established in 2003. These polyesters can be used for fiber manufacturing where a lower viscosity grade is required than for bottle fabrication. However, some challenges arise during these processes. The color of the obtained fibers is often yellowish or grayish . It is not rare that virgin PET has to be added to postconsumer PET for fiber processing. During repeated recycling, chain cleavage of the polyester often takes place which results, even for fiber manufacturing, in too low viscosities . Alternatively, for the usage of fully recycled PET products, even for bottle grade products, PET can be reprocessed with the help of a solid state polycondensation (SSP) process. The SSP process is a thermal treatment to increase the molecular weight of PET to a value which is needed for certain products. It is a postpoly-condensation step under solid conditions. Condensation byproducts like water, ethylene glycol, or low molecular weight oligomers are evaporated during this process. PET is heated to temperatures ranging between the glass transition temperature ([T.sub.g]) and the melting temperature ([T.sub.m]) of around 200[degrees]C-240[degrees]C [3,4]. During the SSP process, a water- and oxygen-free atmosphere (dry inert gas stream or vacuum) has to be ensured [5-7]. Chain cleavage and hydrolysis are prevented in the water- and oxygen-free atmosphere. The SSP process lasts more than 8 hrs. Basically, the SSP process results in improved quality of PET due to an increase of the molecular weight and a decrease of the formation of volatile products. However, the SSP process is expensive due to the high heating times and its process conditions. Furthermore, PET can be subject to the formation of an undesirable gray color during the SSP process which is mainly a result of the presence of antimony catalysts (such as [Sb.sub.2][O.sub.3] or Sb [[OOC--C[H.sub.3]].sub.3] ) which can be reduced to metallic antimony under these conditions [2,9]. As substitute for the SSP process, chain extenders can be used to increase the molecular weight of PET. Chain extenders are multifunctional molecules that link the end groups of polymer segments (e.g., COOH end groups in PET) together which results in a high molecular polymer. Chain extenders can be classified into bi-, tri-, or multifunctional products. While the Afunctional molecules extend PET in a linear way, tri- or higher functional molecules act as crosslinker.
A variety of chain extenders with different kinds of reactions have been developed in the past. Furthermore, the development of chain extenders as polyester masterbatches has been published . The usage of chain extenders in PET has its advantages and disadvantages (listed in Table 1).
Most favorable for PET are addition-type chain extenders. Chain extenders with reactive end groups like bisepoxy compounds, carboxylic dianhydride, and diisocyanates are some examples of addition-type chain extenders [11,12]. However, these compounds may result in branched (bisepoxy compounds, carboxylic dianhydride, and diisocyanates) or less thermally stable products (diisocyanates). Further commonly used addition-type chain extenders are bisoxazolines such as 1,3-phenylene-bis-oxazoline (1,3-PBO). 1,3-PBO is a bifunctional heterocyclic compound in the group of cyclic imino ethers (imidates) which has a general formula of -N=C-O-. The synthesis of oxazolines has been published by several authors [13-20]. Preferably, oxazolines are synthesized from their corresponding nitriles and 2-aminoethanol as published for the first time by Witte and Seelinger (Scheme 1) [21-24], In the case of 1,3-PBO, 1,3-dicyanobenzene (isophthalodinitrile) is the corresponding starting material. As catalysts, mostly zinc salts such as zinc acetate are used as weak Lewis acids [16,20], The generated byproduct N[H.sub.3] has to be removed.
Oxazolines undergo many ring-opening reactions with a variety of functional groups like carboxyl groups [25-27]. Furthermore, polymerizations can be done with aliphatic dicarboxyl acids (e.g., adipic acid or sebacic acid) or aromatic dicarboxyl acids (e.g., terephthalic acid or isophthalic acid) and bisoxazolines [28-33]. Moreover, reactions with oligomers and polymers can be performed with the help of bisoxazolines which are chain extension reactions. Bisoxazolines are commonly used as chain extenders in several polymers with carboxyl end groups such as polyamides [34-37], poly(lactic acid) [38-40], poly(butylene terephthalate) [41,42], and PET [12,34,42-48], In the case of PET, the reaction with the carboxylic acid end groups is the most important one. The reaction of carboxyl acid groups with oxazolines results in the formation of ester amide bonds, and many applications are generated with the aid of this type of reaction [49-52], In Scheme 2, the chain extension reaction of 1,3-PBO with the carboxyl end groups of PET is presented. Just small amounts of 1,3-PBO in the range of 0.3 wt% to 0.7 wt% are needed to get good results with regard to high intrinsic viscosities of PET . The chain extension of PET with 1,3-PBO proceeds in a linear way, which is important in many applications.
Beside chain extenders which react with carboxyl end groups, chain extenders which react with the hydroxyl end groups of PET (e.g., bislactams) have been investigated [53,54]. As an example, N,N'-carbonylbiscaprolactum (CBC) is a commonly known chain extender first published in 1956 by Meyer , Furthermore, Loontjens et al. published studies using bislactams as chain extenders for a variety of polymers such as polyurethanes, polyamides, and polyesters [34,56-62].
CBC is synthesized by the reaction of [epsilon]-caprolactam and phosgene in the presence of a tertiary amine (e.g., triethyl amine) as acid scavenger (Scheme 3).
The chain extension reaction of CBC occurs at terminal hydroxyl or amine groups. It is also possible, that CBC reacts with terminal carboxyl groups, but the reaction with OH or N[H.sub.2] end groups is faster . In the case of PET, the terminal OH groups react with CBC via evaporation of [epsilon]-caprolactam or ring opening reaction (Scheme 4). Linear chain extension is obtained which is very important, especially for fiber production to achieve good spinnability. The chain extension takes place within 3 min and small amounts of 0.1 to 1.0 wt% of chain extender are needed ,
In this work, the impact of 1,3-PBO, CBC, and their combinations on the properties of fiber grade PET is investigated. The application of chain extenders is not very common in polyester fiber spinning. One reason for this is the fact that PET with lower intrinsic viscosities is needed for fiber spinning than for bottle applications. But, if recycled PET is used, the intrinsic viscosity of the PET recyclate may be too low to achieve good spinnability. Research is needed to proof that chain extenders can be used successfully to spin polyester fibers from PET recyclate. The characteristics of the obtained PET are analyzed by viscosity measurements, size exclusion chromatography (SEC), and in particular by parallel plate rheology, and differential scanning calorimetry (DSC).
PET for fiber production was provided by Maerkische Faser GmbH (Premnitz, Germany). The fiber grade polyester with an intrinsic viscosity of 0.63 dL [g.sup.-1] was used for the extrusion experiments with the chain extenders.
The solvent 1,1,1,3,3,3-hexafiuoropropane-2-ol (HFIP) was purchased from Fluorochem (Hadfield, United Kingdom) and chloroform was bought from J. T. Baker (Deventer, The Netherlands). 2,6-Di-tert-butyl-4-mcthylphcnol, o-cresol, and 2-bromobenzoic acid were received from Sigma-Aldrich, (Taufkirchen, Germany). Bromophenol blue was received from Merck (Darmstadt, Germany) and ethanolic potassium hydroxide solution was purchased from Fluka (Taufkirchen, Germany). Furthermore, CBC was supplied from DSM (Geleen, The Netherlands), and 1,3-PBO was provided from Adeka Palmarole (Basel, Switzerland).
First of all, PET was dried at 130[degrees]C in an oven overnight at least for 10 hrs. The Micro 15 cc Twin Screw-Extruder (DSM, Geleen, The Netherlands) was used for the extrusion experiments. The "DSM Xplore Data Acquisition and Control v1.11" Software was used to measure the screw force. About 11 g of PET were molten at 290[degrees]C with 100 rounds per minute screw rotation speed under nitrogen atmosphere. Different amounts of the chain extenders were added to PET and mixed for 5 min in the extruder in a discontinuous process. The applied concentrations of the chain extenders were 0 wt%, 0.1 wt%, 0.2 wt%, 0.3 wt%, 0.5 wt %, 1.0 wt%, and 2.0 wt%. After the compounding process, the samples were ground in a cryomill (6800 Freezer/Mill, SPEX CertiPrep, Stanmore, United Kingdom) to achieve good homogeneity and good solubility for further analysis.
The inherent viscosity ([[eta].sub.inh]) of the polyester was measured to analyze the influence of the chain extenders on the polymer molecular weight. About 0.3300 g of PET was weighed in a 25 mL graduated flask and dissolved in HFIP. The viscosity of this solution was measured in a water bath at 25[degrees]C using an Ubbelohde viscosimeter (type 0a) (Schott AG, Mainz, Germany). The inherent viscosity was calculated according to the following equation:
[[eta].sub.inh] = ln([[eta].sub.rel])/[beta] = ln([eta]/[[eta].sub.0])/[beta] = ln(t/[t.sub.0])/[beta] (1)
where [[eta].sub.inh] is the inherent viscosity and [[eta].sub.rel] the relative viscosity. [beta] is the mass concentration, [eta] is the viscosity of the PET solution, and [[eta].sub.0] is the viscosity of the solvent (here, HFIP). t and [t.sub.0] are the flow times of the PET solution and the solvent HFIP, respectively.
Size Exclusion Chromatography
Molecular weights ([M.sub.n] and [M.sub.w]) and molecular weight distribution (D) of PET samples were determined by SEC. PET was dissolved in HFIP and diluted with chloroform to a volume concentration of chloroform/HFIP 98:2 vol% according to Weisskopf , An HPLC pump (PU-2080plus, Jasco, Tokyo, Japan) equipped with an evaporative light scattering detector (PL-ELS-1000, Polymer Laboratories, Amherst, MA) was used.
2,6-Di-tert-butyl-4-methylphenol (c = 250 mg [mL.sup.-1]) was used as internal standard, and narrow distributed polystyrene standards (PSS Polymer Standards Service GmbH, Mainz, Germany) were used to achieve calibration. One pre-column (8 mm x 50 mm) and four SDplus gel columns (8 mm x 300 mm, MZ Analysentechnik, Mainz, Germany) were applied at a flow rate of 1.0 mL [min.sup.-1] at 20[degrees]C. The separation process was performed on polystyrene/divinylbenzene columns (50, 100, 1,000, and 10,000 [Angstrom], PSS Polymer Standards Service GmbH, Mainz, Germany). Results were evaluated using the PSS WinGPC UniChrom software (Version 8.1.1).
The rheology measurements were performed with a parallel plate rheometer (Discovery HR-3 hybrid rheometer, TA Instruments-Waters L.L.C., New Castle, DE). PET was molten at 290[degrees] C and measured with a gap of 600 [micro]m in a frequency range of 0.1-100 Hz with 2% oscillation. The complex viscosities ([eta]*), storage moduli (G'), and loss moduli (G") were calculated.
Differential Scanning Calorimetry
DSC analyses were carried out using the Netzsch DSC 204 (NETZSCH-Geraetebau GmbH, Selb, Germany). About 10 mg PET were weight into aluminum pans, closed with a lid and pierced. The samples were heated up to 300[degrees]C with a heating rate of 20 K [min.sup.-1] under nitrogen flow and hold there for 10 min to delete the thermal history of the polymer. After that, the pans were cooled down to 200[degrees]C, where the crystallization begins, followed by an isothermal annealing step of 30 min at this temperature. To measure the melting point ([T.sub.m]) and melting enthalpy ([DELTA][H.sub.f]) of PET, the samples were heated up again to 300[degrees]C with a heating rate of 10 K [min.sup.-1], cooled down to 20[degrees]C with a cooling rate of 10 K [min.sup.-1] to measure the crystallization during cooling of the melt ([T.sub.c]) and heated up a third time to 110[degrees]C with a heating rate of 20 K [min.sup.-1] to determine the glass transition temperature ([T.sub.g]).
The crystallinity ([[chi].sub.c]) of the samples was calculated according to the following equation:
[[chi].sub.c] = 100% x [DELTA][H.sub.f]/[DELTA][H.sub.f100%] (2)
[DELTA][H.sub.f100%] is the melting enthalpy of a fully crystallized PET with a value of 140 J [g.sup.-1] [64-66].
Furthermore, the lamellar thickness distribution of the polymers was calculated with the aid of an approach of Hoffman, Davis, and Lauritzen using the Gibbs-Thomson equation (Eq. 3) .
L = 2[sigma][T.sup.0.sub.m]/[DELTA][H.sub.fV]([T.sup.0.sub.m] - [T.sub.m]) (3)
Here, L is the lamellar thickness, [sigma] is the surface free energy (0.106 J [m.sup.-2]), [T.sub.0.sup.m] is the equilibrium melting temperature of an infinite crystal (564 K), [T.sub.m] is the melting temperature, [DELTA][H.sub.fV] is the melting enthalpy per volume unit of a fully crystallized PET (2.1 x [10.sup.8] J [m.sup.-3] at [rho] = 1,455 g [cm.sup.-3]) [65,66,68,69].
Carboxyl End-Group Titration
To determine the carboxyl end groups of PET, titrations with ethanolic potassium hydroxide (KOH) solution with a concentration of 0.05 mol [L.sup.-1] using bromophenol blue as indicator were performed. The titer (t) of the KOH standard solution was determined with dried 2-bromobenzoic acid. Approximately 0.8-1.5 g of PET were dissolved in 20.0 g of o-cresol at 80[degrees]C, quenched with chloroform and titrated against potassium hydroxide standard solution. As blank, 20.0 g of o-cresol mixed with chloroform was also titrated. In each case, triple determinations were performed. The COOH concentration (in mmol [kg.sup.-1]) was calculated according to the following equation:
c(COOH) = [V,(KOH) - [V.sub.0](KOH)] x c(KOH) x t x [10.sup.3]/m(PET) (4)
RESULTS AND DISCUSSION
The chain extenders were typically pre-mixed with PET (dry blend) and added afterwards to the extruder. Furthermore, PET was added in a few samples to the extruder at 290[degrees]C with 100 rounds per minute screw rotation speed and mixed there with chain extenders to get a first indication of chain extension reactions. As an example, the extrusion curve is presented in Fig. 1 where 1,3-PBO was added to the PET melt. The screw force of the extruder reveals that more force is needed to rotate the screws after the addition of the chain extender. An increase of the screw force from 1854 N to 2059 N is measured in that case. This indicates an increase of the melt viscosity and, hence, an increase of the molecular weight of PET. The color of extruded PET samples did not change significantly by addition of small amounts of the chain extenders (up to 0.5 wt%). At higher chain extender concentrations, the melt discolors to yellow upon addition of 1,3-PBO or CBC.
At first, the inherent viscosities of PET materials which were extruded with the addition of chain extenders were determined. In Fig. 2, the inherent viscosities of virgin PET and PET extruded with 1,3-PBO in concentrations of 0, 0.1, 0.2, 0.3, 0.5, 1.0, and 2.0 wt% are presented. First, the inherent viscosity decreases, when virgin PET is extruded without any chain extender which is due to the thermal degradation of PET at 290[degrees]C. Then, 1,3-PBO was added to virgin PET in an extrusion process. Already at small 1,3-PBO concentrations, increases of the inherent viscosities are observed. PET extruded with 0.1 wt% 1,3-PBO has reached the inherent viscosity of virgin PET. Upon addition of higher amounts of 1,3-PBO, the inherent viscosity of PET increases also. A maximum inherent viscosity with 0.66 dL [g.sup.-1] is reached at 1.0 wt%.
Compared to 1,3-PBO, CBC is at low concentrations (0.1 - 0.2 wt%) less effective; however, small effects are also observable (Fig. 3). At a concentration of 0.3 wt%, the inherent viscosity of virgin PET is obtained. At high concentrations, the inherent viscosity increases very strongly up to 0.92 dL [g.sup.-1] at 2.0 wt% CBC.
Size Exclusion Chromatography
Furthermore, SEC measurements of the PET samples were performed in a chloroform/HFIP (98/2 vol%) solution. In Fig. 2, the results of virgin PET and PET extruded with 0.1-2.0 wt% 1,3-PBO are given. While virgin PET has an average molar mass of 14.4 kg [mol.sup.-1] (number average molar mass [M.sub.n]) and 40.7 kg [mol.sup.-1] (weight average molar mass [M.sub.w]) has PET which was extruded with 1,3-PBO has an average molar mass up to 21.1 kg [mol.sup.-1] ([M.sub.n]) and 48.1 kg [mol.sup.-1] ([M.sub.w]). Furthermore, the molar mass is rather broadly distributed. The molecular weight distributions are in the range of 2.3-2.8. However, this virgin PET has also a broad molecular weight distribution of 2.8. The increases of the molar mass and the inherent viscosity of PET indicate that chain extension reactions were successfully performed with addition of 1,3-PBO in a reactive extrusion process at 290[degrees]C.
SEC measurements were also done for PET samples extruded with CBC (0.1-2.0 wt%). In Fig. 3, the SEC results of these samples are presented. These results reveal strong increases of the average molar masses from 14.4 kg [mol.sup.-1] ([M.sub.n]) and 40.7 kg [mol.sup.-1] ([M.sub.w]) up to 25.4 kg [mol.sup.-1] ([M.sub.n]) and 83.6 kg [mol.sup.-1] ([M.sub.w]). Here, the increases of the molar mass and the inherent viscosity of PET show that a reaction of the chain extender CBC with PET was successfully realized in a reactive extrusion process at 290[degrees]C, too.
The rheological characteristics of the PET samples extruded with chain extenders are also determined. With the aid of a parallel plate rheometer, the complex viscosity ([eta]*), storage modulus (G'), and loss modulus (G") are measured.
First, the results of the complex viscosity measurements are presented in Figs. 4 and 5. In Fig. 4, the results of the [eta]* measurements of PET extruded with 1,3-PBO are depicted. It reveals that the PET melts have almost Newtonian-like behavior at angular frequencies [omega] of 0.5-330 rad [s.sup.-1]. At higher shear rates, shear thickening behavior is obtained. This behavior is observed for all samples except for PET extruded with 2.0 wt% 1,3-PBO. In that case, Newtonian-like behavior is observed up to 20 rad [s.sup.-1], and it shows shear thinning behavior at higher angular frequencies. The reason for this effect is the higher molar mass of the PET chains. Due to the high molar mass, the polymer chains are highly entangled and disentanglement occurs at high shear rates. In other cases, the polymer chains are less entangled and get tangled up at high shear rates which lead to this shear thickening behavior. The PET samples which are compounded with 1,3-PBO have larger complex viscosities than the reference sample extruded without added chain extender. For example, at an angular frequency of 0.6 rad [s.sup.-1], the complex viscosity ([eta]*) increases from 75 to 205 Pa s which indicates chain extension induced by 1,3-PBO.
Figure 5 presents the results of the complex viscosity ([eta]*) measurements of PET extruded with CBC. Newtonian-like behavior is also found for the PET samples compounded with small amounts of CBC (0-0.5 wt%) up to [omega] = 330 rad [s.sup.-1]. The polymer chains are strongly entangled at high shear rates resulting in increased complex viscosity. In contrast, PET compounded with 1.0-2.0 wt% CBC shows a direct decrease of the complex viscosity with increase of the angular frequency. Due to the high molar mass, these polymer chains are highly entangled, and they are disentangled at high shear rates. The complex viscosities at small angular frequencies (here, [omega] = 0.6 rad [s.sup.-1]) raised from 75 Pa s (0% CBC) to 342 Pa s (2.0% CBC). This indicates a strong increase of a chain extension by CBC.
Moreover, in the following figures the storage moduli (G') and the loss moduli (G") of PET treated with CBC are presented (Figs. 6 and 7). The storage moduli of PET which was extruded with CBC are shown in Fig. 6. At higher angular frequencies ([omega]), the storage moduli are also higher, because of the higher shear rates of the melts. The high shear rate leads to an increased entanglement of the polymer chains and, as a result, an increased storage modulus. In addition, higher chain extender concentrations reveal also higher storage moduli. For example, at [omega] = 0.6 rad [s.sup.-1] the storage modulus of PET extruded without chain extender is 2.8 Pa and the storage modulus of PET extruded with 2.0 wt% CBC is 49 Pa at [omega] - 0.6 rad [s.sup.-1]. The reason for this is that increased CBC concentrations extend the polymer chains and the entanglement of these chains and, as a result, the storage modulus is also higher.
The loss moduli (G") are also higher at higher angular frequencies ([omega]) as depicted in Fig. 7. Due to the higher shear rate of the melt, the internal friction of the chains increases, too. Thus, the loss of energy which is defined as loss modulus (G") is also larger. An increase of the loss modulus due to the addition of the chain extender to the PET melt is also shown. The loss modulus at [omega] = 0.6 rad [s.sup.-1] rises from 45 Pa (0 wt% CBC) up to 212 Pa (2.0 wt% CBC). The chain extended polymers have higher loss moduli because of increased internal friction due to their longer chains after compounding.
For the processability of polymer materials, the rheology is quite important. As shown earlier, the polymer melt is not affected significantly by the addition of small amounts of chain extenders (0.1-0.5 wt%) compared to PET which was compounded without additives. Thus, polymer processes at 100 Hz (about 630 rad [s.sup.-1]) can be performed with addition of small amounts of 1,3-PBO and CBC.
Differential Scanning Calorimetry
The thermal behavior of the chain extended PET samples was investigated, too. The results of the DSC measurements of PET samples are shown in Table 2. The glass transition point ([T.sub.g]), melting points ([T.sub.m]), melting enthalpy ([DELTA][H.sub.f]), crystallinity ([[chi].sub.c]), and crystallization during cooling of the melt ([T.sub.c]) are listed in Table 2.
In Figs. 8-10, the results of the DSC measurements are presented. In Fig. 8, the melting endotherms of virgin PET and PET extruded with 1,3-PBO with concentrations from 0 to 2.0 wt% are shown. The melting endotherm represents the transition of the solid semi-crystalline structure to the molten, amorphous state. The DSC diagram of the investigated virgin PET which was recorded with a heating rate of 10 K [min.sup.-1] up to 300[degrees]C has a broad endothermic signal with two maxima in the range of 235[degrees]C to 260[degrees]C. The two maxima at 242[degrees]C and 252[degrees]C correspond with two melting endotherms. The multiple melting behavior of PET has been published by different authors [70-73]. In all cases, annealing at 200[degrees]C for 30 min was performed after erasure of the thermal history in the first heating step. After that, the samples were heated up to 300[degrees]C with a heating rate of 10 K [min.sup.-1] to determine the melting characteristics. It is thought that the first melting endotherm is related to the lamellae formed during crystallization and the second melting endotherm corresponds to the larger lamellae generated by recrystallization of the smaller ones which leads to fusion of lamellae [70,72]. But, it is also possible that the second melting endotherm is caused by formation of bigger crystalline sequences from amorphous or partially ordered sequences at the interface of the crystallites. They may undergo an orientation process due to the annealing step .
Here, the lower melting temperature has its maximum at 242[degrees]C, and the higher melting temperature has a maximum at 252[degrees]C. Figure 8 demonstrates that the second melting area decreases with increasing 1,3-PBO content of PET together with an increase of the first melting area. This is most pronounced for PET which was extruded with 1,3-PBO concentrations of [greater than or equal to] 0.5 wt %. The area under the melting endotherms is directly related to the crystallinity of PET. Thus, the reaction of PET with 1,3-PBO disturbs the formation of highly crystalline structures in the polymer resulting in smaller crystalline structures. 1,3-PBO is a foreign building block which is inserted into the polyester chains and reduces, as a result, the crystallinity of the polymer.
In analogy to PET compounded with 1,3-PBO, the higher melting temperature of PET extruded with CBC decreases also at higher CBC concentrations (Fig. 9, Table 2). Here as well, CBC acts as disrupter of the crystalline structure of PET. The DSC diagram of PET compounded with 2.0 wt% of CBC depicts that the higher melting area disappears completely (Fig. 9). This shows that high amounts of CBC have a stronger influence on the crystallinity of PET than 1,3-PBO.
In general, the crystallinity of the PET samples decreases after extrusion with 1,3-PBO and CBC (Fig. 10, Table 2) due to the disturbance of the crystalline structure by these chain extenders. Both chain extenders are foreign building blocks in the polymer chains. The degrees of crystallinity over the whole melt endotherms are shown in Fig. 10. It reveals that high chain extender contents decrease the crystal structure of PET.
Furthermore, the lamellar thickness distribution of these samples was calculated and the results are presented in Fig. 11 (PET compounded with 1,3-PBO) and in Fig. 12 (PET compounded with CBC). These figures show, that the larger crystallites decrease after addition of 1,3-PBO and CBC to PET during extrusion. While virgin PET has a lamellar thickness distribution ranging from 87 to 184 [Angstrom] with maxima at 117 and 145 [Angstrom], PET which was extruded with 2.0 wt% 1,3-PBO has a lamellar thickness distribution of 82 to 170 A and maxima at 118 and 136 [Angstrom] (Fig. 11). The lamellae of PET extruded with CBC are even smaller. The lamellae distribution of PET extruded with 2.0 wt% CBC is in the range of 87 [Angstrom] to 161 [Angstrom] and has only one maximum at 118 [Angstrom] (Fig. 12). These results demonstrate that the chain extenders disturb the formation of crystallites. In addition, in the case of PET compounded with 1,3-PBO the lamellar thickness decreases already at small amounts of 1,3-PBO (cf. Fig. 11). The formation of lamellae is promoted by the hydrogen bonds between the end groups of different PET chains. As 1,3-PBO reacts with the COOH end groups of PET, less hydrogen bonds and, as a result. less lamellae are formed. However, the addition of small amounts of CBC during extrusion of PET has less influence on the lamellar thickness.
Carboxyl End-Group Titration
To validate that 1,3-PBO reacted with the carboxyl end groups, COOH end-group titrations were performed. PET extruded without any chain extender was at first titrated. The amount of COOH end groups in PET extruded without chain extenders was about 38.9 [+ or -] 0.4 mmol [kg.sup.-1]. Compared to this, the COOH concentration of PET extruded with 1.0 wt% CBC decreased slightly to 33.1 [+ or -] 0.1 mmol [kg.sup.-1]. A stronger decrease is obtained for PET which was extruded with 1.0 wt% of 1,3-PBO for which a COOH concentration of 22.9 [+ or -]0.3 mmol [kg.sup.-1] was determined. These results lead to the conclusion that the COOH end group concentration decreases after addition of 1,3-PBO. This influences the lamellar thickness distribution as mentioned before.
Combination of 1,3-PBO and CBC
Finally, 1,3-PBO and CBC were combined in the extrusion process with PET. In Fig. 13, the results of the inherent viscosity measurements are presented. It is obvious that a combination of these two chain extenders results in an additional effect. Except in the case of PET extruded with 0.1 wt% 1,3-PBO and 1.0 wt% CBC, higher inherent viscosity results are reached compared to the application of only one chain extender (cf. Figs. 2 and 3). For example, PET extruded with 0.1 wt% 1,3-PBO and 0.3 wt% CBC has an inherent viscosity of 0.65 dL [g.sup.-1], whereas for PET extruded with 0.3 wt% CBC an inherent viscosity of 0.63 dL [g.sup.-1] was measured (cf. Fig. 3). A combination of these two chain extenders is beneficial because of their different reactions with the terminal end groups of PET. 1,3-PBO reacts with the COOH terminal groups of PET and CBC has higher reactivity with the OH end groups of PET (cf. Schemes 2 and 4).
High molar mass PET can be obtained by the addition of chain extenders such as 1,3-PBO and CBC. The chain extenders are linearly linked to the COOH and/or OH terminal groups of PET. The addition of small amounts (0.1-2.0 wt%) of 1,3-PBO and CBC to the PET melt in the extrusion process results in an increase of the inherent viscosity which corroborates with increased molar mass which was, furthermore, proven by SEC. The chain extension reactions occur fast in a reactive extrusion process at 290[degrees]C which is a one-step process that can be of interest for application during fiber production from postconsumer PET. Factors like the reactivity of the chain extenders with the end groups of PET also play an important role for their effectiveness.
An increase of the storage moduli (G'), loss moduli (G"), and complex viscosities ([eta]*) was observed in all cases for PET which was compounded with chain extender addition. After addition of 2.0 wt% CBC, the complex viscosity of PET increased up to 267 Pa s as measured by rheology at 0.6 rad [s.sup.-1] and 290[degrees]C. Also an increase of 46 Pa for the storage modulus and of 167 Pa for the loss modulus was measured at 0.6 rad [s.sup.-1] and 290[degrees]C. However, for PET which was extruded with small amounts of the studied chain extenders (0.1-0.5 wt%), no marked changes in its rheological properties were determined. The higher the amount of chain extender is, the lower are the angular frequencies, where the Newtonian-like behavior ends. Newtonian-like behavior is observed up to 330 rad [s.sup.-1] for PET extruded without chain extender and up to 20 rad [s.sup.-1] for PET extruded with 2.0 wt% 1,3-PBO. In the case of PET which was extruded with 2.0 wt% CBC, non-Newtonian-like behavior was observed. Generally, the extended PET chains are higher entangled and, thus, the disentanglement starts at lower shear rates which results in non-Newtonian behavior. For stable spinning of fibers Newtonian behavior or at least relatively constant viscosity at the spinning speeds is required. Thus, the application of high amounts of chain extenders results in unstable fiber spinning.
The thermal properties of PET are not markedly affected by the addition of small amounts of the investigated chain extenders to the polymer melt in the extrusion process. However, the crystallinity and the lamellar thickness distribution of PET decrease after addition of chain extenders to the melt. Especially, the secondary crystallization is disturbed after addition of chain extenders. Small concentrations of the chain extenders (0.2 wt% 1,3-PBO or 0.3 wt % CBC) do not have significant effects on the thermal properties and crystallization of PET.
The COOH terminal group titrations show a slight decrease of the COOH content of PET compounded with CBC and a strong decrease of the COOH content of PET compounded with 1,3-PBO. 1,3-PBO has higher reactivity with the COOH terminal groups of PET than CBC.
Moreover, an additional effect was observed by combining both chain extenders in a reactive extrusion process. It has been shown that 1,3-PBO and CBC are very effective chain extenders for PET and can be used to replace the SSP process for a variety of applications.
The authors would like to thank Landesamt fur Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (LANUV), European Union (European Regional Development Fund; ERDF) and Effizienz-Agentur Nordrhein-Westfalen (EfA) for supporting the research project ResPoSe (funding code: 21060227612). Parts of the analytical investigations were performed at the Center for Chemical Polymer Technology CPT, which was supported by the EU Commission and the federal state of North Rhine-Westphalia (Grant no. 300088302). Furthermore, the authors thank the project partners ADVANSA GmbH, Hamm, Germany, and Umweltdienste Kedenburg GmbH, Beckum, Germany, for support and helpful discussions and all DWI colleagues for their assistance.
[1.] A. Nicastro and D. Koehlmann, Kunststoffe, 10, 78 (2016).
[2.] D. Berg, K. Schaefer, A. Koemer, R. Kaufmann, W. Tillmann, and M. Moeller, Macromol. Mater. Eng., 301, 1454 (2016).
[3.] F. Awaja and D. Pavel, Eur. Polym. J., 41, 1453 (2005).
[4.] S.N. Vouyiouka, E.K. Karakatsani, and C.D. Papaspyrides, Prog. Polym. Sci., 30, 10 (2005).
[5.] Q. Gao, N.-X. Huang, Z.-L. Tang, and L. Gerking, Chem. Eng. Sci., 52, 371 (1997).
[6.] S.A. Cruz and M.J. Zanin, Appl. Polym. Sci., 99, 2117 (2006).
[7.] U.K. Thiele, Polyester Bottle Resins Production, Processing, Properties and Recycling, PETplanet Print, Heidelberg, Germany (2007).
[8.] S.B. Maerov, J. Polym. Sci. Pol. Chem., 17, 4033 (1979).
[9.] S.M. Aharoni, Polym. Eng. Sci., 38, 1039 (1998).
[10.] D. Berg, K. Schaefer, and M. Moeller, Macromol. Sym., 375, 1600180, 1 (2017).
[11.] A.J. Dijkstra and J.A.W. Reid, U.S. Patent 3,553,157 A (1971).
[12.] H. Inata and S. Matsumura, J. Appl. Polym. Sci., 30, 3325 (1985).
[13.] R.H. Wiley and L.L. Bennett, Chem. Rev., 44, 447 (1949).
[14.] W. Seeliger, E. Aufderhaar, W. Diepers, R. Feinauer, R. Nehring, W. Thier, and H. Hellmann, Angew. Chem., 78. 913 (1966).
[15.] J.A. Frump, Chem. Rev., 71, 483 (1971).
[16.] S. Kobayashi, Prog. Polym. Sci., 15, 751 (1990).
[17.] K. Aoi and M. Okada, Prog. Polym. Sci., 21, 151 (1996).
[18.] P. Garg, S. Chaudhary, and M.D. Milton, J. Org. Chem., 79, 8668 (2014).
[19.] M. Trose, F. Lazreg, M. Lesieur, and C.S. Cazin, J. Org. Chem., 80, 9910(2015).
[20.] C. Lixin and X. Yahong, J. Mater. Sci. Lett., 22, 953 (2003).
[21.] H. Witte and W. Seeliger, Angew. Chem. Int. Ed., 11, 287 (1972).
[22.] H. Witte and W. Seeliger, Liebigs Ann. Chem., 1974, 996 (1974).
[23.] S. Kobayashi and T. Saegusa, Encyclopedia of Polymer Science and Technology, John Wiley & Sons, New York, NY (2011).
[24.] K. Neubold, DE Patent 2135644 A (1973).
[25.] W. Seeliger, E. Aufderhaar, W. Diepers, R. Feinauer, R. Nehring, W. Thier, and H. Hellmann, Angew. Chem. Int. Ed., 5, 875 (1966).
[26.] B.M. Culbertson, Prog. Polym. Sci., 27, 579 (2002).
[27.] C. Zuerbig, "Bifunktionelle cyclische Imidsaureester ais Additive flir die Mehrfachverarbeitung von Polyethylenterephthalat." Ph.D. Thesis. RWTH Aachen University, Aachen, Germany (1997).
[28.] A. Jaeger, DE Patent 1050540 A (1959).
[29.] Y. Sano, J. Polym. Sci. Pol. Chem., 27, 2749 (1989).
[30.] J. Luston, J. Kronek, O. Markus, I. Janigova, and F. Bohrne, Polym. Advan. Technol., 18, 165 (2007).
[31.] J. Luston, J. Kronek, and I. Janigova, J. Macromol. Sci., Part A, 47,716 (2010).
[32.] T. Kagiya, S. Narisawa, T. Maeda, and K. Fukui, J. Polym. Sci. Pol. Lett., 4, 257 (1966).
[33.] F. Kenichi, K. Tsutomu, K. Mikio, M. Taneo, and N. Shizuo, U.S. Patent 3,476,712 A (1969).
[34.] T. Loontjens, K. Pauwels, F. Derks, M. Neilen, C.K. Sham, and M. Serne, J. Appl. Polym. Set., 65, 1813 (1997).
[35.] Y. Chalamet and M. Taha, J. Polym. Sci. Pol. Chem., 35, 3697 (1997).
[36.] Y. Chalamet, M. Taha, and B. Vergnes, Polym. Eng. Sci., 40, 263 (2000).
[37.] Y. Chalamet, M. Taha, F. Berzin, and B. Vergnes, Polym. Eng. Sci., 42, 2317 (2002).
[38.] J. Kylma, J. Tuominen, A. Helminen, and J. Seppala, Polymer, 42, 3333 (2001).
[39.] J. Tuominen, J. Kylma, and J. Seppala, Polymer, 43, 3 (2002).
[40.] P.V. Bonsignore, U.S. Patent 5,470,944A (1995).
[41.] S.L. Liu, W.G. Li, B.R. Liang, and Y.H. Yu, Eur. Polym. J., 36, 2159 (2000).
[42.] A. Douhi and A. Fradet, J. Polym. Sci. Pol. Chem., 33, 691 (1995).
[43.] N. Torres, J.J. Robin, and B. Boutevin, J. Appl. Polym. Sci., 79, 1816 (2001).
[44.] N. Cardi, R. Po, G. Giannotta, E. Occhiello, F. Garbassi, and G. Messina, J. Appl. Polym. Sci., 50, 1501 (1993).
[45.] H. Inata and S. Matsumura, J. Appl. Polym. Sci., 32, 5193 (1986).
[46.] H. Inata and S. Matsumura, J. Appl. Polym. Sci., 33, 3069 (1987).
[47.] G.P. Karayannidis and E.A. Psalida, J. Appl. Polym. Sci., 77, 2206 (2000).
[48.] C.R. Nascimento, C. Azuma, R. Bretas, M. Farah, and M. L. Dias, J. Appl. Polym. Sci., 115, 3177 (2010).
[49.] P. Mueller, C. Woerner, and R. Muelhaupt, Macromol. Chem. Phys., 196, 1917 (1995).
[50.] J. Luston, F. Boehme, H. Komber, and G. Pompe, J. Appl. Polym. Sci., 72, 1047 (1999).
[51.] J. Luston, F. Bohme, H. Komber, and G. Pompe, J. Macromol. Sci. Pure, A35, 1045 (1998).
[52.] N. Chau, S. Matsuda, and Y. Iwakura, Makromol. Chem., 180. 1435 (1979).
[53.] H. Inata and S. Matsumura. J. Appl. Polym. Sci., 32, 4581 (1986).
[54.] H. Inata and S. Matsumura, J. Appl. Polym. Sci., 34, 2609 (1987).
[55.] H.R. Meyer, "Ueber die Umsetzungsprodukte von Caprolactam und Polycaprolactam mit Formaldehyd und Phosgen," Ph.D. Thesis, ETH Zuerich, Zuerich, Switzerland (1956).
[56.] J.A. Loontjens, F.J.M. Derks, and C.K. Sham, WO Patent 1996034909 A1 (1996).
[57.] J.A. Loontjens and B.J.M. Plum, WO Patent 1998047940 Al (1998).
[58.] J.A. Loontjens and B.J.M. Plum, WO Patent 2001066633 Al (2001).
[59.] S. Maier, T. Loontjens, B. Scholtens, and R. Miilhaupt. Angew. Chem. Int. Ed., 42, 5094 (2003).
[60.] J.A. Loontjens, "Performance Materials by a Modular Approach," Ph.D. Thesis, Technical University Eindhoven, Eindhoven, The Netherlands (2005).
[61.] J. Loontjens, U.S. Patent 20070161773 Al (2007).
[62.] J. Zimmermann, T. Loontjens, B.J. Scholtens, and R. Miilhaupt, Biomaterials, 25, 2713 (2004).
[63.] K. Weisskopf, J. Polym. Sci. Pol. Cliem., 26, 1919 (1988).
[64.] B.H. Bimestre and C. Saron, Mater. Res.-Ibero-Am. J., 15, 467 (2012).
[65.] A. Mehta, U. Gaur, and B. Wunderlich, J. Polym. Sci. Pol. Phys., 16, 289 (1978).
[66.] B. Wunderlich, Macromolecular Physics: Crystal Melting, Elsevier Science, Amsterdam, Netherlands, and New York, USA (2013).
[67.] J.D. Hoffman, G.T. Davis, and J.I. Lauritzen, in Treatise on Solid State Chemistry, Vol. 3, N.B. Hannay, Ed., Springer, Boston, MA (1976).
[68.] X.F. Lu and J.N. Hay, Polymer, 42, 9423 (2001).
[69.] J.D. Badia, E. Stromberg, S. Karlsson, and A. Ribes-Greus, Polym. Degrad. Stabil., 97. 98 (2012).
[70.] Y. Kong and J.N. Hay, Polymer, 44. 623 (2003).
[71.] C.X. Zhou and S.B. Clough, Polym. Eng. Sci., 28, 65 (1988).
[72.] P.J. Holdsworth and A. Turner-Jones, Polymer, 12, 195 (1971).
[73.] Z.G. Wang, B.S. Hsiao, B.B. Sauer, and W.G. Kampert, Polymer, 40, 4615 (1999).
[74.] G. Groeninckx and H. Reynaers, J. Polym. Sci. Pol. Phys., 18, 1325 (1980).
Dennis Berg, Karola Schaefer [iD], Martin Moeller [iD]
DWI--Leibniz Institute for Interactive Materials e. V. and Institute of Technical and Macromolecular Chemistry (ITMC), RWTH Aachen University, Aachen, Germany
Correspondence to: K. Schaefer; e-mail: email@example.com or M. Moeller; e-mail: firstname.lastname@example.org
Contract grant sponsor: Landesamt fur Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (LANUV); Contract grant number: 21060227612. Contract grant sponsor: State of North-Rhine Westfalia of Germany; Contract grant number: 300088302. Contract grant sponsor: European Regional Development Fund; Contract grant number: 21060227612.
Caption: SCHEME. 1. Synthesis of 1,3-PBO [17,21,221.
Caption: SCHEME. 2. Reaction of 1,3-PBO with the carboxyl end groups of PET [12,26,32].
Caption: SCHEME. 3. Synthesis of CBC by the reaction of e-caprolactam and phosgene .
Caption: SCHEME. 4. Potential reactions of CBC and hydroxyl end groups of PET [10,601. (a [red]) two substitutions of [epsilon]-caprolactam, (b [black]) combination of substitution of [epsilon]-caprolactam and ring opening reaction, (c [blue]) two ring opening reactions. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 1. Extrusion curve of PET with addition of 0.3% 1,3-PBO measured at 290[degrees]C with 100 rpm.
Caption: FIG. 2. SEC results and inherent viscosities ([[eta].sub.Inh]) (orange hatched) of virgin PET compounded with 1,3-PBO (0.1-2.0 wt%) at 290[degrees]C in comparison to virgin and extruded PET (without chain extender). The number average molar mass ([M.sub.n]) (black) and weight average molar mass ([M.sub.w]) (red) are presented. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 3. SEC results and inherent viscosities ([[eta].sub.Inh]) (orange hatched) of virgin PET compounded with CBC (0.1-2.0 wt%) at 290[degrees]C in comparison to virgin PET and extruded PET (without chain extender). The number average molar mass ([M.sub.n]) (black) and weight average molar mass ([M.sub.w]) (red) are presented. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 4. Complex viscosity ([eta]*) versus angular frequency ([omega]) of PET extruded with 1,3-PBO (concentration ranging from 0 to 2.0 wt%) measured at 290[degrees]C. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 5. Complex viscosity ([eta]*) versus angular frequency ([omega]) of PET extruded with CBC (concentration ranging from 0 to 2.0 wt%) measured at 290[degrees]C. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 6. Storage moduli (G') versus angular frequency ([omega]) measured at 290[degrees]C of PET extruded with CBC (concentration ranging from 0 to 2.0 wt%). [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 7. Loss moduli (G") versus angular frequency ([omega]) measured at 290[degrees]C of PET extruded with CBC (concentration ranging from 0 to 2.0 wt%). [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 8. DSC diagrams of virgin PET and PET extruded with 1,3-PBO in concentrations of 0-2.0 wt%. The melting enthalpy is depicted. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 9. DSC diagrams of virgin PET and PET extruded with CBC in concentrations of 0-2.0 wt%. The melting enthalpy is presented. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 11. Lamellar thickness distribution (L) of virgin PET and PET extruded with 1,3-PBO (concentration from 0 to 2.0 wt%).
Caption: FIG. 12. Lamellar thickness distribution (L) of virgin PET and PET extruded with CBC (concentration from 0 to 2.0 wt%).
TABLE 1. Advantages and disadvantages using chain extenders in a reactive extrusion process instead of the SSP process of PET. Advantages Disadvantages Faster than SSP process Impact on the properties of PET (crystallinity, mechanical properties) Easy application Byproducts may be generated (reactive extrusion) in small scales Addition of small Decomposition of amounts are required chain extenders may occur Chain extenders may be harmful (reactive end groups) TABLE 2. DSC results of virgin PET and PET compounds which were extruded without and with different amounts of the chain extenders CBC or 1,3-PBO. PET sample [T.sub.g] [DELTA] [T.sub.m/Peak/1] ([degrees]C) [H.sub.f] ([degrees]C) (J [g.sup.-1]) Virgin 82.3 39.0 242.4 0% chain extender 84.2 36.9 242.8 0.1% 1,3-PBO 83.2 38.5 242.9 0.2% 1,3-PBO 84.5 36.7 242.8 0.3% 1,3-PBO 84.1 37.1 242.6 0.5% 1,3-PBO 83.6 34.2 243.0 1.0% 1,3-PBO 84.4 32.9 242.9 2.0% 1,3-PBO 84.2 31.9 242.7 0.1% CBC 84.4 37.2 243.2 0.2% CBC 84.2 37.4 243.1 0.3% CBC 83.5 38.4 242.8 0.5% CBC 83.8 36.7 243.1 1.0% CBC 84.2 34.2 242.5 2.0% CBC 83.4 30.0 242.5 PET sample [T.sub.m/Peak/2] [T.sub.cc] [[chi].sub.c] ([degrees]C) ([degrees]C) (%) Virgin 251.9 186.7 27.8 0% chain extender 251.9 183.9 26.4 0.1% 1,3-PBO 251.5 192.4 27.5 0.2% 1,3-PBO 251.4 185.4 26.2 0.3% 1,3-PBO 251.0 194.8 26.5 0.5% 1,3-PBO 250.6 192.3 24.5 1.0% 1,3-PBO 249.3 192.4 23.5 2.0% 1,3-PBO 248.9 185.5 22.8 0.1% CBC 252.2 187.7 26.6 0.2% CBC 251.7 186.3 26.7 0.3% CBC 251.7 188.9 27.4 0.5% CBC 251.6 187.2 26.2 1.0% CBC 249.8 186.2 24.4 2.0% CBC -- 192.1 21.4 FIG. 10. Degree of crystallinity of virgin PET and PET extruded with chain extenders (1,3-PBO or CBC) in concentrations of 0-2.0 wt%. [Color figure can be viewed at wileyonlinelibrary.com] [[chi].sub.C]/% 1,3-PBO CBC virgin 27.8 27.8 0% 26.4 26.4 0.1% 27.5 26.6 0.2% 26.2 26.7 0.3% 26.5 27.4 0.5% 24.5 26.2 1.0% 23.5 24.4 2.0% 22.8 21.4 Note: Table made from bar graph. FIG. 13. Results of inherent viscosity measurements of virgin PET compounded with 1,3-PBO and CBC in different concentrations in comparison to virgin and extruded PET (without chain extender) (compare with the results given in Figs. 2 and 3). [Color figure can be viewed at wileyonlinelibrary.com] [[eta].sub.inh]/dL x [g.sup.-1] 1,3-PBO + CBC virgin 0.63 0.0% 0.61 0.1% 1,3-PBO + 0.1% CBC 0.62 0.3% 1,3-PBO + 0.3% CBC 0.64 0.3% 1,3-PBO + 0.1% CBC 0.64 0.1% 1,3-PBO + 0.3% CBC 0.65 1.0% 1,3-PBO + 1.0% CBC 0.66 0.1% 1,3-PBO + 1.0% CBC 0.68 1.0% 1,3-PBO + 1.0% CBC 0.74 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Author:||Berg, Dennis; Schaefer, Karola; Moeller, Martin|
|Publication:||Polymer Engineering and Science|
|Article Type:||Technical report|
|Date:||Feb 1, 2019|
|Previous Article:||Effects of Silica Nanoparticles on Tribology Performance of Poly(Epoxy Resin-Bismaleimide)-Based Nanocomposites.|
|Next Article:||Residual Wall Thickness of Water-Powered Projectile-Assisted Injection Molding Pipes.|