Size-dependent Effect of MgAl-Layered Double Hydroxides Derived from Mg(OH)2 on Thermal Stability of Poly(vinyl chloride)

: Developing green thermal stabilizer for poly(vinyl chloride) (PVC) today is a great challenge for materials as polymer filler. Here, the “salt-oxide/hydroxide” route in mild conditions is used to fabricate a series of Mg 2 Al-CO 3 -LDH samples from Mg(OH) 2 precursors with different average particle sizes from 202 ± 10 nm to 334 ± 13 nm. A linear correlation is observed for the lateral size of the platelets between Mg 2 Al-CO 3 -LDH samples and their associated Mg(OH) 2 precursors. After surface-organo-modification (SOM), organophilic Mg 2 Al-CO 3 -LDH samples are found to be highly dispersed into PVC and investigated as environment-friendly thermal stabilizers. From the static/dynamic tests, the performances are strongly enhanced and related to the particle size of the LDH stabilizer, with the yellowish color aspect appearing later than for the commercial HT-3/PVC. Among the LDH series, the platelets with an average particle lateral size of about 220 ± 10 nm perform the best for the thermal stability for PVC polymer. Among the series, the corresponding PVC composite film presents comparatively the minimum color value in static/dynamic discoloration test, exhibiting the longer ignition time for proton initial release as well as the longer stability time in dehydrochlorination test. It underlines that the salt-oxide/hydroxide route is an efficient and environmentally friendly process in producing high-performance green LDH stabilizer for PVC.


Introduction
Poly(vinyl chloride) (PVC) is the third-most widely used polymer around the world, and largely applied in various fields such as construction, chemical, and packaging, electric and electronic products, and so forth [1][2][3].In addition, the statistics anticipate a global demand for PVC to reach 46.3 million tons in 2019 and to continuously increase [4].However, PVC resins have a poor thermal stability: this is mostly due to the fact that particularly, its processing is occurring at temperature higher than the dehydrochlorination and thermal degradation temperature thus making it highly fragile.Therefore, a thermal stabilizer is usually required to add during the processing of PVC to enhance its thermal stability as well as to retard its thermal degradation [5].Hitherto, various thermal stabilizers have been developed and generally used in the market, for instance, lead salts, metal soaps, organotin, and other organic composites [6].
However, most of them are toxic towards human being in a certain degree.Indeed, the lead salts have been strictly limited by the European Union's RoHS and REACH directives.Usually, more than one organic stabilizer is needed to stabilize thermally PVC, moreover they are mostly hazardous chemicals.To challenge high-performance and environment-friendly thermal stabilizers for PVC, the layered double hydroxides (LDH) appear promising.Indeed, for the last forty years and since the first patent (US4299759) in the 1980's [7], LDHtype materials have drawn more and more attention as a potential "green" thermal stabilizer for PVC [8][9][10][11].The thermal stability mechanism of LDHs is found to proceed in two ways: one is the anion-exchange reaction between Cl -in HCl from PVC and CO 3  2-into the LDH interlayer region, and the other is the possible neutralization reaction between HCl produced during PVC degradation and the alkaline LDH layers [12].The first is to trap the aggressive Cl -anions, LDH host is acting as a scavenger, while in the second LDH is dissolved under pH, thus buffering the medium.Most interestingly, LDH filler is acting under pH and chloride anions concentration stimuli, and silent otherwise.This aspect of a dormant load until reacting when necessary is reminiscent of LDH filler used in polymer coatings to inhibit corrosion phenomenon occurring onto metal substrate [13].In decades, the effects of different Mg/Al ratios, intralayer cations and interlayer anions substitution of LDHs on the thermal stability property of PVC have been investigated in detail in the literature [9-11,14], but t it is here beyond the scope of the study.
Since the thermal stability performance of the LDH/PVC composite should also depend on the state of dispersion of the LDH particles within the composite, the focus is on their particle size and size distribution [15][16][17][18][19][20].Usually, LDH-type materials were prepared by the so-called coprecipitation from the reaction between metal salts in alkaline medium as reported in academy as well as in industry.However, in this route, huge amount of water is produced in the washing steps that makes irrelevant in terms of sustainable development at any large scale-up [21][22][23][24][25].
More recently, the so-called "green" synthesis route has been developed using mixture between metal oxide and hydroxide or salt-oxide and hydroxide to reduce drastically the amount of water used during the washing step [26][27][28][29][30].However, the former is difficult to yield LDH platelets suitable as thermal stabilizer for transparent PVC due to their sizes too large beyond micrometer [31], while the latter is not reported to fabricate LDH filler as PVC stabilizer so far [20].
Therefore, there is still a great challenge to fabricate high-performance and environment-friendly LDH thermal stabilizer for PVC composites.
In this work, a series of Mg 2 Al-CO 3 -LDH samples with different particle sizes was first prepared by the salt-hydroxide route using Mg(OH) 2 , Al(NO 3 ) 3 •9H 2 O, and light magnesium carbonate (LMC) as the feeding materials and by adjusting the particle size of Mg(OH) 2 precursor.In a second step, the LDH filler was organo-modified onto the platelets, leading to a surface compatibilized hybrid structure so called hereafter surface organo-modified LDH, SOM-LDH, towards polymer.Finally, the PVC/SOM-LDH composites were melt-blended in an open twin-wheel mill, and subsequently investigated for their thermal stability performance as a function of the LDH particle size.Finally, the precipitate was centrifuged and washed with water for 1 cycle.The obtained slurry was separated in two parts: one was for the preparation of the organic-modified Mg 2 Al-CO 3 -LDH; and the other was for further physical characterization.The LDH samples were noted as LDH from a to g, corresponding to Mg(OH) 2 precursor from 0 h to 16 h.

Organic modification of Mg 2 Al-CO 3 -LDH
In order to enhance the compatibility with the PVC, the LDH samples were further organicmodified by stearic acid as described in the literature [32][33][34].For example, the LDH-b slurry (31.20 g, solid content = 18.7 %, 12 mmol) was dispersed in 500 mL of water (solid content < 2 %), and then transferred into a 1 L three-neck flask under mechanical stirring.Subsequently, 0.47 g (1.65 mmol) stearic acid was dissolved into 50 mL of hot water, and then placed into the above-mentioned flask when the reaction temperature reached 80 °C.The resulting slurry was reacted at 80 °C for 2 h.Finally, the suspension was centrifuged and dried at 60 °C overnight leading to the surface organic-modified LDH (SOM-LDH) powder.

Preparation of SOM-LDH/PVC composites
A series of SOM-LDH/PVC composites were prepared by melting blending method.Typically, in the static thermal stability test, 100.00 g PVC resin, 2.00 g stabilizers (0.44 g Ca(st) 2 , 0.44 g Zn(st) 2 and 1.12 g SOM-LDHs) and 40.00 g DOP were mixed in a beaker.The above mixtures were plasticized in an open twin-wheel mill (BP-8175-A, Baopin Ltd., China) at 36 rpm and 30 rpm roll speeds at 185 °C for 4 min, and then the gap between the two rolls was controlled at 0.4 mm to prepare transparent SOM-LDH/PVC composite films, respectively.In the dynamic thermal stability test, a blend of 100.00 g PVC resin, 1.00 g stabilizers (0.22 g Ca(st) 2 , 0.22 g Zn(st) 2 and 0.56 g SOM-LDHs) and 25.00 g DOP was pressed by twin-wheel mill at 190 °C to obtain transparent SOM-LDH/PVC film.In both cases, a neat PVC film was fabricated as the blank sample for comparison without a SOM-LDH filler using the same mass of other stabilizer and DOP as for SOM-LDH/PVC composite.

Characterization
Powder X-ray diffraction (PXRD) patterns were recorded on a Shimadzu XRD-6000 X-ray diffractometer (40 kV, 30 mA, Cu Ka radiation, λ= 0.154 nm) in the range from 3 ° to 70 °/2θ with a scan step of 10 °/min.The morphology and particle sizes of the samples were measured using a Zeiss Supra 55 scanning electron microscope (SEM) and the average particle size was calculated from more than 100 particles for each sample.Fourier transform infrared (FT-IR) spectra in the range of 4000 to 400 cm -1 with 2 cm -1 resolution were collected on a Bruker Vector 22 infrared spectrophotometer using the KBr disk method with a ratio of sample/KBr of 1: 100 and the film method for SOM-CO 3 -LDH/PVC films.The thickness of the LDH samples and the dispersion of LDHs in PVC resins were estimated using a Digital Instruments Version 6.12 atomic force microscope (AFM).Differential scanning calorimeter (DSC) measurements were carried out in N 2 gas flow using a differential scanning calorimeter (DSC, METTLER TOLEDO, DSC1 STARe System).

Transparent test of SOM-LDH/PVC composites
100.00 g PVC, 8.00 g stabilizers (1.76 g Ca(st) 2 , 1.76 g Zn(st) 2 and 4.48 g LDHs), and 50.00 g DOP were placed into a beaker and mixed uniformly.The mixture was plasticized at 185 °C by twin-wheel mill, and then hot-pressed and moulded with a thickness of 5 mm using a plate vulcanizing machine.A series of Mg 2 Al-CO 3 -LDH samples were synthesized by the salt-hydroxide method using the above-prepared Mg(OH) 2 precursors, Al(NO 3 ) 3 and LMC as the feeding materials.Fig. 1 shows the PXRD patterns of the seven resulting Mg 2 Al-CO 3 -LDH samples.All the samples exhibit the typical PXRD patterns of LDH with the well-defined Bragg reflection peaks, e.g.

Structural analysis of Mg 2 Al-CO 3 -LDH
(003), (006), ( 009) and (110) as marked in the graph.The basal d-spacing of Mg 2 Al-CO 3 -LDH is 0.78 nm calculated from the first three Bragg reflection peaks, which is in agreement with that of CO 3 -type LDH reported in the literature [36].from Kyowa Chemical Industries of Japan.The morphology of the as-made LDH samples and the HT-3 is for all platelet-like structures but with some slight differences: the as-made series exhibit a rather homogeneous hexagonal shape similar to Mg(OH) 2 precursor when lying flat on the substrate with the observation of thin-layer when exposed perpendicular.The reference HT-3 presents circular roundish aspect.Besides, Fig. 3a displays the particle size variation of Mg(OH) 2 and LDH extracted from Figs. S2 (Supporting information) and 2 as a function of the aging time of Mg(OH) 2 .Interestingly, both series present a similar trend with a first rapid growth up to 8 h then to remain constant after.Indeed, Fig. 3b further demonstrates a close linear relationship in terms of particle size between LDHs and Mg(OH) 2 precursors.For example, the LDH platelets with the smaller particle lateral size of about 202 ± 10 nm were prepared from the smaller particles of Mg(OH) 2 presenting a lateral size of 49 ± 5 nm, while the larger particle size of 334 ± 13 nm was from larger particles of Mg(OH) 2 of 176 ± 10 nm.This linear dependence may suggest a topochemical reaction between both materials, reminiscent of previous observation for the transformation from single to double layered hydroxides [37].An in-situ characterization beyond the scope of the study will be necessary to unravel the reason of such lateral size dependence between the reactant and the reaction product, i.e., between the brucite and the LDH phase and why this LDH series mirrors that of the precursors.The above results suggest that it is an available and efficient route to prepare LDHs with different particle sizes by controlling that of Mg(OH) 2 precursor employed in the salt-hydroxide route.The particle size of HT-3 is 219 nm, close to that of LDH when using Mg(OH) 2 -1h.Interestingly, the AFM image and associated height profile in Fig. 3c, d show that the platelets thickness of LDH-b sample is ca.3.5 nm consisting of ca. 4 layers of LDH nanosheets, which is much thinner than ca.24 nm measured for HT-3 (Fig. 2h) and as usually obtained by the traditional coprecipitation [38,39].Besides, all the samples from LDH-a to -g display a stacking of platelets of similar thickness, that is close to that observed from samples prepared by another green route using MgO, Al 2 O 3 and NaNO 3 [40].
Intuitively, thinner LDH platelets should improve the dispersion state of the LDH particles into PVC, thus to be more exposed to the polymer chains and more efficient to capture HCl when PVC degrades [41][42][43][44].Evidently, the lateral size may play a role as a barrier effect, the LDH platelets acting as a sacrificial reservoir in improving PVC stability.Quantitatively an increase from 202 ± 10 nm to 334 ± 13 nm shifts the time of a "yellow" appearance from 50 min to 30 min, however, they are all black after 80 min.In addition, Fig. 5b further demonstrates the color values of ΔE as a function of thermal degradation time.For example, SOM-LDH-b/PVC composite presents the minimum color value before turning black after 80 min, showing the best thermal stability among the series.In the dynamic test at 190 °C, Fig. 5c and d show that both SOM-LDH-a and -b/PVC exhibit the same thermal stability than that with HT-3/PVC, but in the first 30 min.The performance in the series after 45 min of test is directly ranked by the lateral size variation.One may note that the particle lateral size of the asmade SOM-stabilizers plays a key role in improving the thermal stability of PVC.As intuitively thought, an appropriate particle size of the filler favours its dispersion state into the composite, then being more exposed, it is more efficient to protect PVC from its degradation effect.To sum up, a small thickness and particle lateral size of LDH platelets have both significant influence on the thermal stability of LDH/PVC composites.The former is controlled during the "green" synthesis, the later can be tuned by reaction conditions.During the thermal degradation process, a certain amount of HCl is released from PVC composites.Fig. 6a shows the concentration of protons as a function of the degradation time up to 240 min.In comparison, all the prepared SOM-LDH samples exhibit much higher capacity to inhibit the HCl release from the PVC degradation than neat PVC or HT-3/PVC.The detailed results are scrutinized in terms of the ignition time (t 1 ) and the stability time (t 2 ) for deeper comparison (Fig. 6b).As reported in Ref. 45, t 1 is the time for the protons (H + ) to be released in solution and t 2 represents the time to reach a concentration of 2*10 -3 mol/L.For example, t 1 is rather short of 9 min for neat PVC, and 30 min for HT-3/PVC, while much longer times are observed for the as-made SOM-LDH/PVC.Comparatively, t 1 ranks as follows: neat PVC < HT- In the series SOM-LDH/PVC composites, SOM-LDH-b/PVC exhibits the higher thermal stability associated to the longer ignition time t 1 of 114 min, and the longer stability time t 2 of 234 min.These above results demonstrate that the SOM-LDH is highly efficient to stabilize PVC in temperature, and to endow the polymer with the performance of interest for application.

Thermal degradation behaviour of SOM-LDH/PVC composites
Besides, transparency property is another important parameter for PVC product.

Conclusions
In this work, the "salt-oxide/hydroxide" route in mild conditions is adopted to fabricate a series of Mg 2 Al-CO 3 -LDH samples with different particle sizes from 202 ± 10 nm to 334 ± 13 nm from Mg(OH) 2 precursor that is prepared by the SNAS method.The prepared Mg 2 Al-CO 3 -LDH samples are surface-organo-modified to render them organophilic and easily dispersible into PVC.As thermal stabilizer, the as-made series exhibits higher performance to inhibit the PVC thermal degradation when compared to commercially available LDH standard.Using static/dynamic tests, it is demonstrated that enhanced performances are strongly related to the particle size of the LDH stabilizer.Among the investigated LDH samples in our work, the better candidate comparatively into the series presents an average particle lateral size of about 220 ± 10 nm, close to the commercial standard.Computational modeling should be carried out to better understand such lateral size dependence against performance to further optimize PVC stability.
In summary, the salt-oxide/hydroxide green route appears as transferable and environmentally friendly process to produce high-performance green LDH stabilizer for PVC.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

2 where
Static thermal stability of SOM-LDH/PVC composites were carried out by discoloration test and dehydrochlorination test.Discoloration test: The discoloration test of PVC sheets was performed according to ISO 305: 1990-4 standard.The PVC film was approximately cut into 1×1 cm sheets and heated in temperature-controlled oven at 195 °C in air.Strips were removed from the oven every 10 min.The thermal stability of SOM-LDH/PVC composites was appraised by the color differences using a Konica Minolta CR-400 type automatic color difference meter.The △L*, △A* and △ B* values were determined by this color meter.The CIE-L*A*B* values could be used to formulate a value for total change as △E by the equation: △E = (△ ) △L*, △A* and △B* values represented the white (+) and black (-), red (+) and green (-), yellow (+) and blue (-), respectively.The △E value was chosen to describe the discoloration of PVC sheets.A low △E value corresponds to a low color difference.Dehydrochlorination test: The release rate of HCl gas produced by the PVC decomposition was estimated by a dehydrochlorination test based on ISO 182-2: 1990-12[14,35].1.50 g of transparent PVC film were cut into small pieces by 1×1 mm.Fig.S1shows the schematic diagram of dehydrochlorination test equipment.The PVC pieces were placed in a 50 mL threeflask and immersed in an oil bath at 180 °C.The HCl gas released from PVC pieces was recovered into a 100 mL deionized water at room temperature by N 2 gas.The pH value of the water was measured by a Mettler Toledo FE28 type pH meter (resolution, 0.01) and the concentration of H + in water was calculated.Dynamic thermal stability of SOM-LDH/PVC composites was performed by discoloration test under the plastication process with twin-wheel mill at 190 °C.A piece of PVC film was taken out of the twin-wheel mill every 5 min and their color changes were measured by the abovementioned automatic color difference meter.

Fig. 2
Fig. 2 displays SEM images of the seven Mg 2 Al-CO 3 -LDH samples obtained by the salt-

Fig. 5
Fig. 5 shows the color change of SOM-LDH/PVC composite sheets in the static (a and b) and

Fig. 5 .
Fig. 5. Results of discoloration (a, c) and color values (b, d) of transparent PVC sheets at 195 °C

3 /
PVC < SOM-LDH-a/PVC < SOM-LDH-e/PVC < SOM-LDH-d/PVC < SOM-LDH-c/PVC < SOM-LDH-b/PVC, and for t 2 neat PVC < HT-3/PVC < SOM-LDH-a/PVC <SOM-LDH-e/PVC < SOM-LDH-c/PVC < SOM-LDH-d/PVC <SOM-LDH-b/PVC.In addition, Fig. S6 displays first derivative curve of the HCl release rate as a function of time for the PVC samples according to data from Fig. 6a, showing that the ignition of HCl release for neat PVC is much faster than for PVC samples with filler as well as comparing the release rate from the relative intensity of the derivative curves.The observed trend agrees well with the discoloration test results in Fig. 5.
Fig. 6d displays the DSC curves of LDH/PVC composite films.The glass transition temperature

Fig, 6 .
Fig, 6.(a) HCl release curves for the PVC samples heated at 180 °C; (b) The induction time and