Structural and thermal degradation behaviour of reclaimed clay nano-reinforced low-density polyethylene nanocomposites

Lupolen 1800 S (trade name of LDPE) polymer was supplied by Northern Polymers and Plastics Ltd., UK. It has a melting point of 106 °C and a V-2 rating in accordance with UL 94 (vertical burning test) at 1.6 mm thickness. The spent oil based mud (OBM) was donated by a local oil and gas company. Montmorillonite, K10 powder was purchased from Sigma-Aldrich, UK and used as reference material in sample analysis.

Filler was recovered from OBM waste in a series of thermal treatments. The first treatment involved heating the OBM at 50 °C for 12 h to eliminate as much of the petroleum fraction as possible. The temperature was then increased to 80 °C for a further 12 h. Recovery of the solid content was enhanced by decanting oil-water mixture from the top of the sample. In the second stage of recovery, the furnace temperature was increased to 700 °C and held for a further 12 h to vaporise the remaining heavy oil fractions. Only the solid content of the OBM waste was recovered. The recovered mud was milled to a powder using ball mill (Ultra Turrax, IKA).

LDPE/OBMFs nanocomposites were manufactured by using a commercially available LDPE and recovered OBM clay. The filler concentrations used in this study were are 2.5 wt%, 5.0 wt%, 7.5 wt% and 10.0 wt%. LDPE Polymer in pellet form and OBMFs were dried at 90 °C overnight prior to melt compounding. Melt compounding of different wt% of concentrations of OBMFs with LDPE was carried out using TwinTech Extrusion Ltd. 10 mm twin screw extruder at 60 rpm over five heating zones: 1st zone (120 °C), 2nd zone (200 °C), 3rd zone (210 °C), 4th zone (200 °C) and die/5th zone (200 °C). The compounded granules were prepared by using a pelletiser which were then injection moulded into a bar mould (dual cavity) for different analyses using a temperature of 230 °C with a moulding pressure of 10 bar. The samples were then left to cool to room temperature before submitting to different analysis and characterisation processes.

To observe the morphology of OBMFs and their dispersion characteristics in LDPE polymer matricesx, a cryogenic sampling technique was has been followed for sample preparation. The fractured samples were gold coated using sputter deposition for 2 min prior to the experiment. The fractured sample sections were observed using a Zeiss EVO LS10 Scanning Electron Microscope (SEM) with a magnification of 250x, 4.5 mm working distance and 25.00 kV accelerating potential. To determine the approximate elemental composition of the samples, energy dispersive x-ray analysis (EDXA) (Oxford Instruments INCA Energy) was carried out. The morphology of cryo-fractured samples were analysed using SEM and EDXA. Attenuated Total Reflectance-Fourier Transform Infra-Red spectroscopy was performed using a Thermo Scientific Nicolet iS10 FTIR Spectrometer setting for 32 scans between 4000 and 400 cm⁻¹. The influence of water vapour and carbon dioxide in air was minimised by subtracting peaks from blank measurement.

To identify the degradation and decomposition behaviour of neat LDPE and LDPE/OBMFs nanocomposites, Thermo-Gravimetric Analysis (TGA) was carried out using a TA TGA Q500 instrument. This analysis represented the weight variation of a measured sample on ramp mode from room temperature (31.1 °C) to 1000 °C at a rate of 10 °C per minute. To ascertain the glass transition, melting and crystallisation phases of the materials, Differential Scanning Calorimetry (DSC) was carried out using a TA Q100 instrument by adopting the heat-cool-heat method. The principle of this method is to determine the heat exchange difference between the sample and the reference pan (an empty aluminium pan). The temperature was set from −20 °C to 250 °C at a rate of 10 °C per minute.

XRD analysis was carried out on a Siemens D5000 diffractometer with Cu Kα radiation (λ = 0.15418 nm) and 0.05° 2θ step size. The scan angle covered the range from 3° to 60° 2θ for layer materials and nanocomposites at a scan speed of 2° min⁻¹. The dispersion morphology of the nanocomposites was measured by Transmission electron microscopy (TEM), obtained on a JEOL JEM2010, 200 kV. An 80 nm thickness sample was prepared by using a Leica Ultracut-UCT microtome without any staining.

Dynamic shear measurements were obtained with a rheometer (AR 1000, TA Instruments) using a parallel plate geometry (8 mm diameter and 3.1 mm gap). Samples were tested to identify the viscoelastic properties as a function of temperature. The temperature range used in this study was varied from 0 to 90 °C, with a heating rate of 3 °C/min the samples were scanned at a fixed frequency of 1 Hz, with a strain of 0.2%.

Tensile and flexural (3-point bending) tests were carried out using Instron 3382 universal testing machine connected with Bluehill 3 software. The load was measured using a 100kN load cell. Tensile experiment was conducted using the dog-bone shaped samples which were prepared as per ASTM method D-638 and flexural experiment was conducted according to The European Standard EN ISO 14125. All tests were conducted at constant strain rate of 2 mm/min and the data presented here is the average of 5 samples.

The ATR-FTIR spectrum analysis of neat LDPE and its nanocomposite has been carried out and the resulting spectra are presented in Fig. 1.

A summary of ATR-FTIR peak assignments is presented in Table 1 based on information published in the literature [16, 33‐36].

The chemical structure of LDPE/OBMFs nanocomposites has been identified by using ATR-FTIR which shows the IR transmittance peaks at 2914 cm⁻¹ and 2850 cm⁻¹ corresponding to CH₂ stretching, the peaks at 1465 cm⁻¹ and 718 cm⁻¹ representing to the bending and rocking of CH₂, respectively. The peaks at 1377 cm⁻¹ highlight the presence of interlayer carbonate ions. The peak at 1082 cm⁻¹ represents the stretching of Si-O, and the peaks at 640 cm⁻¹, 630 cm⁻¹ and 608 cm⁻¹ correspond to the oxide bonds of metals such as Si, Al, Mg etc. The peak at 580–525 cm⁻¹ represents the crystalline forms of rare earth oxides which is observed in LDPE nanocomposite samples with 5.0, 7.5 and 10.0 wt% OBMFs content. It can articulate that the peaks at lower bands between 640 and 540 cm⁻¹ correspond to the chemical structure formed by different minerals present in the OBMFs nanoclay.

To identify the elemental composition and explain the addition of OBMFs in the LDPE polymer matrix, Energy Dispersive X-ray (EDXA) analysis was carried out as presented in Fig. 2a–e.

Figures 2a–e show the EDX spectra and also the elemental composition of neat LDPE, LDPE with 2.5 wt% OBMFs, LDPE with 5.0 wt% OBMFs, LDPE with 7.5 wt% OBMFs and LDPE with 10.0 wt% OBMFs nanocomposite. Elemental analysis of this filler has been reported in a Naxos 2018 conference proceedings which showed the presence of O, Na, Al, Si, S, Cl, K, Ca, Fe and Ba. The EDXA study here also confirmed the presence of different elements exist in clay minerals distributed in the LDPE matrix, as highlighted in Fig. 2. The lower limit of detection of EDXA is 0.01 wt%.

To identify the surface topography of reclaimed clay particles from oil based mud waste and LDPE/OBMFs nanocomposite, morphological studies were carried out by using SEM. The SEM micrographs were used to analyse the changes in morphological structure of OBMFs in LDPE matrix.

Figure 3a shows the presence of tightly stacked clay platelets with size ranges up to 1000 nm. Figure 3b shows the cavity and uneven surfaces due to the cryogenic fracture of the sample. However, Fig. 3c shows that there are loosely dispersed OBMFs in the LDPE matrix which has higher concentration of OBMFs, noticeable in the LDPE matrix in Fig. 3d. Figure 3e shows a mix of agglomeration of OBMFs and local scattered filler in LDPE matrix indicating intercalation of clay platelets. However, Fig. 3f shows poor dispersion of OBMFs with a remarkable agglomeration of OBMFs in the LDPE matrix.

The XRD analyses are illustrated in Fig. 4 addressing the diffractograms at (a) wide angle X-ray diffraction (WAXD); (b) small angle X-ray diffraction; (c) comparing XRD patterns between MMT (as a reference) and OBMFs and (d) illustrating a few extra peaks for the LDPE polymer matrix potentially caused by adding these phases through OBMFs addition to the polymer matrix.

In Fig. 4, XRD patterns of LDPE and its nanocomposites are reported. A clear shift of the diffraction peaks of the planes (001) of OBMFs towards lower angles for the LDPE/OBMFs nanocomposites is noted. However, it can be seen that the basal spacing of OBMFs increases with different nanocomposites in different ratios. The d₀₀₁ spacings were calculated using Bragg’s law nλ = 2dsinθ, where λ is the wavelength of X-ray radiation used in the experiment, d corresponds to the distance between diffraction lattice planes and θ is the half of the diffraction angle. Analysing different peaks from these XRD data, it can be highlighted that the basal spacing increases with the increase of filler contents in nanocomposites. The diffraction peak of OBMFs was observed at 2θ = 11.40° which corresponds to a d-spacing of 7.75 Å. The d-spacings of LDPE with 2.5 wt% and 5.0 wt% OBMFs nanocomposites were identified at 9.6° and 5.6° which corresponds to the value of 9.02 Å and 15.77 Å, respectively. Moreover the d-spacings of LDPE with 7.5 wt% and 10.0 wt% OBMFs nanocomposites were determined as 9.06 Å and 16.06 Å.

Analysing the d-spacing data of the corresponding peaks of reclaimed clay and LDPE/OBMFs nanocomposites, it can be highlighted here that the trend of d-spacings among OBMFs and LDPE/OBMFs nanocomposites are not consistent. However, the peak positions of LDPE with 5.0 wt% and 10.0 wt% OBMFs nanocomposites are noticeable. In both of these materials, the peaks of the planes (001) are shifted towards lower angles, which corresponds to the highest d-spacing value of these materials. This finding also agreed with the findings of the SEM (in Fig. 3) study. Based on this SEM study, it is important to mention here that LDPE with 10.0 wt% OBMFs nanocomposites indicates both local exfoliation and also shows agglomeration of clay minerals in the morphology study. Considering the data obtained in this XRD analysis, it can be inferred that OBMFs is exfoliated in LDPE with 5.0 wt% and apparently OBMFs might locally exfoliated in the LDPE polymer matrix. Transmission electron microscopy analysis is believed to clarify the nanomorphology in more detail in the next section.

TEM analyses were carried out to validate the results obtained from thermal and morphology analyses. It can be highlighted in Fig. 5a that the clay plates are barely connected with each other, representing the exfoliation nature of the clay platelets which is also noticeable in Fig. 5b. Figure 5b clearly shows the tendency of dispersion characteristic from exfoliation towards intercalation. It can be referred as the exfoliation-intercalation interphase.

In Fig. 5c, the loose but structured network of clay platelets is visible which indicates the intercalation nature of these platelets. However, the dark shades and spots in Fig. 5c highlight the agglomeration of platelets, although a few light spots indicate the local exfoliation of clay platelets.

Non-isothermal DSC studies were conducted to elucidate the influence of novel filler OBMFs on the thermal degradation behaviour of nanocomposite materials. Different phases of degradation stages are shown in Fig. 6.

Analysing the DSC thermograms in Fig. 6a, it can be highlighted that there is not any significant change in the glass transition temperature of these nanocomposite materials compared to that of neat LDPE polymer. There is a minor change in glass transition temperature (less than 1 °C) in LDPE with 2.5 wt% OBMFs nanocomposite. However, the melting point remains almost unchanged, only a fraction of a degree Celsius change is noticeable in Fig. 6b. Figure 6c shows a consistent higher recrystallization temperature compare to that of LDPE. These three main heat flow curves in thermograms are discussed in more detail, addressing the potential different thermal properties possessed by these materials.

There is a distinctive peak present in neat LDPE and its nanocomposites in Fig. 6b which represents the melting behaviour of alpha phase crystals associated with the materials. There is not any associated heat curve shoulder noticeable corresponding to the absence of other crystalline phases or the other phases might be very weak compared to the presence of alpha phase in the materials. Considering the thermograms in melting peaks representing the heat capacity value of LDPE and its nanocomposites, the % of crystallinity can be identified using the following equation:

Where ∆Hm is the heat of melting, ∆Hc the heat of cold crystallisation which is 0 as not present in this experiment (∆Hc = 0 in this case), and ∆Hm^° is a reference value if the polymer were 100% crystalline. All the units are in J/g and the value of ∆Hm^° is 293 J/g [37]. The % of crystallinity value of LDPE and its nanocomposites are presented in Table 3.

It is clearly evident that there is a decreasing trend of % of crystallinity as the percentage of filler content increases. The decreasing trend in the % of crystallinity degree of the nanocomposite materials can be explained by the presence of filler contents in polymer matrix, which can hinder the motion of the polymer chain segments and inhibit crystal growth.

The specific heat capacity value of LDPE and its nanocomposites have been identified by analysing the thermograms in Fig. 6c. The specific heat capacity value can be determined by the following equations:

Where Cp is the heat capacity in Joules per Kelvin (JK⁻¹), Q is heat energy in Joule and T is the temperature denoted as °C or K. δQ/δt represents the heat flow and δt/δT corresponds to reciprocal heating rate. By using these two equations, the analysed specific heat capacities of neat LDPE and its nanocomposites are identified, as presented in Table 2.

The heat capacity data presented in Table 2 shows the negative effects of OBMFs in polymer matrix. However, LDPE with 2.5 and 7.5 wt% OBMFs show higher heat capacity property in nanocomposites, LDPE with 10.0 wt% shows the lowest heat capacity indicating a superior thermal conductivity property of this material among different nanocomposites. The heat capacity value decreases about 21% in LDPE with 10.0 wt% OBMFs nanocomposite.

In addition to the SEM observation, it is also important to understand that different interphases exist between LDPE and layered silicate nanocomposites. It is believed that there are two interphases present in LDPE polymer matrix and layered silicates in OBMFs. One of these interphases represents the interphases between the crystal fraction and amorphous fraction of the polymer matrix and the other is between filler and polymer matrix [38]. To understand the thermal degradation behaviour of any nanocomposite material, it is important to identify the different phases present in the polymer matrix predominantly crystalline and amorphous phases exist in the material. Analysing the heat capacity value, Cp in glass transition temperature, mobile amorphous fraction (MAF) can be identified by the following equation:

Where ∆C_p/∆C_{p amp} are the heat capacity increments at the glass transition temperature of LDPE and its nanocomposites and the pure amorphous LDPE polymer, respectively. Using the MAF value, rigid amorphous fraction (RAF) can also be identified by the following equation:

Using the eqs. (4) and (5), the MAF and RAF values are identified and presented in Table 2.

It can be highlighted here that although RAF is non-crystalline, it does not participate in the glass transition due to the parts of the molecules are fixed due to the immobilisation of molecules in the polymer chain [38, 39]. The crystalline fraction in polymer chain, inorganic based filler (OBMFs) (assumed mostly in crystalline phase) and RAF incurs immobilisation in the polymer chain which is evident in the data presented in Table 2. It is also manifested based on the data presented in Table 2 that the RAF value is increased more than 100% in nanocomposites compared to that of neat LDPE. The reason can be explained that the incremental filler content acts as a nucleating agent and increases the nucleation ratio.

Thermal degradation of LDPE/OBMFs nanocomposites has been studied under a dry nitrogen environment using a TA instrument TGA Q500.

The degradation scenario of these materials at different heating stages are analysed and the key findings are presented in Table 3.

The onset degradation temperature of these materials are identified at weight % losses of 5% and 50%. In both cases, LDPE with 10 wt% OBMFs shows the highest onset degradation temperature among the nanocomposites with different wt% filler contents which represents LDPE with 10 wt% OBMFs nanocomposite possessing the highest thermal stability. However, the thermal stability property is significantly affected in LDPE with 7.5 wt% OBMFs. Considering the lowest temperature at D-half (50% weight loss), the presence of peak assignments representing amorphous phase in FTIR analysis and heat capacity value from DSC analysis, it can be inferred here that LDPE with 7.5 wt% OBMFs possesses a good heat capacity property. It is also noticeable that the addition of OBMFs into LDPE matrix significantly reduces the thermal stability of the material based on T_D5% and T_D50% data.

However, the degradation trend is distinctly noticeable as it does not show any significant changes of weight loss % until 300 °C. Degradation mainly occurred between 300 and 500 °C and the poor thermal stability of nanocomposites is clearly visible except for LDPE with 10 wt% OBMFs nanocomposite. In Fig. 7c, LDPE with 10 wt% OBMFs shows the longest time needed to reach the D-half point which also indicates superior thermal stability property among the samples. It is also manifested in this figure that at this D-half point LDPE with 10 wt% OBMFs shows the highest temperature which gives an indication of improving of thermal conductivity property of this material. It is also verified with the lowest heat capacity value obtained from DSC study. It is articulated here that the incremental filler contents may have a strong influence on the thermal stability of LDPE/OBMFs nanocomposites. Filler contents up to 10 wt% reduce the thermal stability of the materials whereas LDPE with 10 wt% possesses the best thermal stability and thermal conductivity properties among the nanocomposites with different wt% filler content.

The storage modulus (E’) represents the load bearing capacity of a material. The effect of OBMFs in influencing this storage modulus is graphically enumerated in Fig. 8a. It is noticeable from Fig. 8a that there was a significant decrease in the storage modulus of LDPE/OBMFs nanocomposites compared to the storage modulus of neat LDPE matrix. This is perhaps due to the decrease in the stiffness of LDPE matrix with the reinforcing effect of the fillers prone to decrease in stress transfer at the interfacial surface. As noticeable in Fig. 8a, the storage modulus decreased drastically with the increase in temperature and the degree of storage modulus drop was significant in the temperature regions between 20 and 70 °C.

The loss modulus curves of neat LDPE and its nanocomposites are presented in Fig. 8b which shows a relaxation peak at 50 °C in neat LDPE decreased to lower temperatures for nanocomposites such as the peaks were at 25 °C, 10 °C, 5 °C and 15 °C for LDPE with 2.5, 5.0, 7.5 and 10.0 wt% OBMFs respectively. There is a significant changes in loss modulus noticeable in the temperature regions between 40 to 55 °C which clearly shows the loss modulus is higher for LDPE with 2.5 and 5.0 wt% OBMFs compared to that of neat LDPE. However, the loss modulus decreased in LDPE with 7.5 and 10.0 wt% OBMFs in those temperature regions. This trend is also clearly shown between 70 to 90 °C temperature regions. This trend completely agreed with the tensile results mentioned in the next section. The equipment which we used for rheology analysis (TA AR1000) is associated with an open hot plate which is not environmentally controlled and we assume the temperature reading in the software may be not the same as the actual temperature in the sample due to the heat absorption by the samples and heat dissipation from the hot plate to open air. It is noticeable that loss modulus reduced drastically from 40 °C which we assume is nearly in the same room temperature the tensile tests were performed at 25 °C.

The ratio of loss modulus to storage modulus is measured which is graphically presented in Fig. 8c to identify the damping properties of the materials. The damping peak in the nanocomposites showed an increased magnitude of tan δ in comparison to neat LDPE. This graph represents the balance between the elastic phase and viscous phase in the polymeric structure. It is noticeable the maximum peak of neat LDPE at 50 °C which is assumed Tg of neat LDPE shifted significantly to lower temperatures for LDPE/OBMFs nanocomposites (Fig. 9).

The mechanical properties of LDPE/OBMFs nanocomposites are presented in Table 4.

It can be seen that the addition of 10.0 wt% OBMFs can improve tensile strength and elongation at break by 17.5% and 13.7% respectively. However the tensile strength and elongation at break of other nanocomposites are remained almost same as tensile strength and elongation at break of neat LDPE. There is an increasing trend of improving tensile modulus among nanocomposites, but only LDPE with 7.5 wt% and 10.0 wt% nanocomposites show higher tensile modulus than that of neat LDPE.

The improvement in mechanical properties of Polyethylene/clay nanocomposites were reported by using compatibilisers by different researchers [40‐42]. However, in our recent studies on LDPE/OBMFs nanocomposites, it is clearly noticeable that there is a little but consistent improvement in tensile property among LDPE/OBMFs nanocomposites. A slight decrease in the tensile strength and modulus can be observed when the OBMFs loading increases from 0 to 5.0 wt%. The significant enhancement in the tensile strength and modulus were obtained with 7.5 wt% and 10.0 wt% OBMFs loading. In addition the decrease trend of elongation at break may be attributed to a reduction in deformability of the rigid interface between nanofillers and matrix [43]. The OBMFs used in this study are nanoparticle in considering size and shape, but observing the micrographs in Fig. 5 it can be highlighted that without adding any grafting compatibilisers, the OBMFs nanoparticles are uniformly dispersed and no agglomeration is noticeable in LDPE with 2.5 wt% and 5.0 wt% fillers loading which are difficult to visualise in SEM images in Fig. 3. However, the orientation of nanoparticles due to the OBMFs compatibility facilitate to act this nanoparticle in a similar manner as fibre which is highlighted in Fig. 5e. After observing carefully the TEM micrographs in Fig. 5a–d, schematic diagrams are presented in Fig. 5e considering the applied force in tensile test and the direction in nanoparticles orientation. It is noticeable that the gradient between nanoparticle orientation and crack propagation plane is decreasing with the increase of filler contents in LDPE matrix which may be attributed to an improvement in materials strength predominantly the fracture strength.