Comprehensive Investigation of the Behavior of Polyurethane Foams Based on Conventional Polyol and Oligo-Ester-Ether-Diol from Waste Poly(ethylene terephthalate): Fireproof Performances, Thermal Stabilities, and Physicomechanical Properties.

The chemical recycling of postconsumer poly(ethylene terephthalate) (PET) bottles to produce highly thermally stable polyurethane foam (r-PUF) with excellent flame-retardant (FR) performance could be applied on an industrial scale to create a sustainable recycling industry. The advantage of oligo-ester-ether-diol obtained from waste PET glycolysis is its application in r-PUF, generating a durable foam with excellent fire resistance at rather low loadings of phosphorus-nitrogen FRs (P-N FRs), especially in high moisture environments. Compared to polyurethane foam from commercial polyol (c-PUF), r-PUF is notably more thermally stable and efficient in terms of flame retardancy, even without adding FRs. By incorporating 15 php diammonium phosphate (DAP) as a P-N FR, r-PUF/DAP self-extinguished 5 s after the removal of the 2nd flame application with a limited oxygen index value of 24%. However, for c-PUF, a much higher DAP (30 php) loading did not exhibit any rating in the vertical burning test. The aromatic moiety in the oligo-ester-ether-diol structure strongly enhanced the compressive strength and thermal stability. The positive outcomes of this study also confirmed that the r-PUF/DAP prepared from oligo-ester-ether-diol not only satisfied the fire safety requirements of polymer applications but also contained a high percentage of postconsumer PET, which could help reduce the amount of recycled polymer materials and improve waste management.

Introduction

Polyurethane foams (PUFs) are reported as being one of the most versatile engineering polymeric foams, which are widely used in a broad variety of high-performance applications, ranging from acoustic insulation materials used in transportation to thermal and electrical insulation materials used in refrigeration technology, construction and building industries, furniture, and appliances, because of the lack of difficulty in handling and combination of exclusive properties, such as their light weight, low density, low thermal conductivity, low moisture absorption, high energy-absorbing ability, and great physicomechanical properties.[1−3]

In view of the increasing demand for PUFs, polymer scientists are exploring the replacement of petrochemical polyols with green, sustainable, and biorenewable polyols, such as vegetable oils and lignocellulosic biomass, which is currently under consideration. This ongoing movement could solve growing concerns caused by the utilization of petroleum feedstocks in polyol production, such as high energy consumption, environmental concerns, global warming resulting from uncontrolled human activities, depletion of fossil fuel reserves, and fluctuation in oil prices. Among the bioresource options, vegetable oils are one of the most promising contestants and are readily available for the production of biobased polyols. Traditional biobased PUF formulations have been prepared by the reaction between diisocyanates and vegetable oil-based polyols from soybean oil and castor oil,[4,5] rapeseed oil,[6] and sunflower oil.[7] Recently, liquid polyols obtained from biomass feedstocks such as cork[8] and bamboo,[9] along with polyols from vegetable oils, have attracted great attention as suitable renewable materials for the production of biobased PUFs.

In addition, instead of using polyols from bioresources, the chemical recycling of polymers is also an effective way to replace fossil fuels for the chemical production of polyols with polyols recycled from plastic waste. Therefore, the proportion of virgin petrochemical-based resins that accumulate in the environment and/or end up being incinerated can be reduced with chemical recycling. Poly(ethylene terephthalate) (PET) is a semicrystalline polyester with high toughness, high strength, light weight, and transparency that is widely used to manufacture textiles, photographic and X-ray films, disposable beverage and soft-drink bottles, as well as materials for food packaging. PET does not directly affect the environment, but due to its massive volume of thermoplastics produced by the irresponsible utilization by society and its nonbiodegradability and chemical resistance, PET could be considered the most recyclable polymer among all thermoplastics.[10] Therefore, there is an increasing demand for the development of recycling techniques and technologies, such as mechanical and chemical recycling, to successfully get PET waste back to its original state and/or generate valuable raw materials from which a new product can be made. Among these recycling methods, chemical recycling is the most effective and acceptable approach to totally/partially decompose PET wastes into monomers/oligomers for the formation of value-added polymers.[11,12] The glycolytic depolymerization of PET is the most important chemical technique, in which glycols, such as ethylene glycol (EG), diethylene glycol (DEG), propylene glycol (PG), glycerol, or their mixtures, are commonly used as glycolysis reagents. Interestingly, in view of the glycolysis reaction, the transesterification reaction degraded PET into monomers for the production of virgin PET as well as promoted the formation of α,ω-dihydroxy terminal materials (polyols), which can be further reacted with other chemicals to generate unsaturated polyester resin,[13−16] alkyd resin,[17] and PUFs.[18] Recently, there have been a few studies utilizing aromatic polyester polyols from glycolysis of PET to fabricate PUF. Aiga Ivdre et al.[19] used a mixture of polyols from depolymerization of PET bottles using PG and from rapeseed oil to produce PUF. It was pointed out that the compression strength and compression modulus of the prepared PUF were higher than those of PUF obtained from commercially available polyols. Dang et al.[20] have published a comprehensive work about recycling of postconsumer PET bottles into PUF. The authors used DEG as a glycolysis reactant and crude glycerol as a trifunctional additive. The results showed that the highest values of the compressive strength, density, and thermal conductivity were obtained at 15 w % loading of crude glycerol.

PUFs are highly flammable materials because of their large surface area, open-cell foam structure, and elemental composition (carbon, hydrogen, oxygen, and nitrogen). During the burning process, combustible PUFs release a large amount of heat and toxic smoke/gases, such as carbon monoxide, hydrogen cyanides, and isocyanates, which have numerous adverse effects on human health and lead to many environmental concerns when exposed to a flaming resource.[21] Therefore, incorporation of flame-retardant (FR) additives into PUFs is the driving force toward the enhancements of fire-resistant behaviors as well as thermal stability to meet special requirements and opens a wide range of applications that are currently obstructed because of the lack of anti-fire safety in the case of neat PUFs. Traditionally, FRs are mainly composed of halogenated additives, which make PUFs good FRs. However, these FRs are very harmful to people’s health and the environment because these halogen-based compounds will release corrosive and toxic smoke during burning. As a consequence, the development of nonhalogen and eco-friendly FRs is a recent research direction in the world. Thus, numerous research studies toward good fire-resistance PUFs by using halogen-free FRs in single use or a combination have been carried out.[22−24]

As halogen-free FRs, organo-phosphorus and phosphorus–nitrogen compounds exhibit a mode of FR action in the gas phase and/or condensed phase, indicating their outstanding fire protection. For example, we have already used triphenylphosphate (TPP) as a phosphorus-containing FR and aluminum trihydroxide (ATH), a commonly used metal hydroxide FR, to enhance the fire retardancy of rigid PUF fabricated from an oligo-ester-ether-diol, a new product of the glycolysis of PET wastes.[25] However, TPP and ATH were added with a rather high loading (25–125 php) into r-PUF to efficiently obtain a fireproof performance, corresponding to the achievement of the UL-94 V-0 ranking. Mengjuan Li et al.[18] studied the FR rigid PUF prepared from a bis(2-hydroxyethyl) terephthalate as a diol, a product of chemical recycling of PET waste from textile sources using EG and dimethyl methylphosphonate (DMMP) as a phosphorus FR. The result showed that the limited oxygen index (LOI) value of PUF was increased to 27.69% with the loading of 18 php DMMP. Recently, phosphorus–nitrogen-containing compounds have been identified as newly effective candidates for FRs that satisfy the specific requirements for anti-fire safety at low loading and govern the critical factors in environmental issues. Phosphorus–nitrogen FRs produced less toxic smoke when they were burning and presented higher thermal stability than phosphorus compounds alone.

The FR mechanism of phosphorus–nitrogen compounds generally acts in the gas phase through the formation of active free radicals (PO•, PO2•, and HPO•) that play an important role in trapping hydrogen and hydroxyl radicals (H• and OH•). These FRs also promote the generation of protective residual char on the material surface during pyrolysis, which prevents heat, oxygen, and flammable gases from attacking the interior polymer.

As a member of the phosphorus–nitrogen FR family, DAP has been used as an effective halogen-free FR for many inherently flammable polymers as well as their composites.[26−28] Like many phosphorous–nitrogen FRs, DAP achieves its fire-resistance through the synergistic effect of a physical mode and a chemical mode (Figure ). Fuel dilution and cooling effects contributed to slowing down heat and fuel-transfer processes during polymer combustion, while phosphorus active free radicals that transferred from phosphorus derivatives could deactivate H• and OH• radicals to cause combustion. In addition, the acceleration of char formation achieved via pyrophosphoric acid and other phosphorus-containing residual chars also contributes to enhancing the flame retardancy of the polymer materials. Taking advantage of its good compatibility, low toxicity, and good fire-resistance performance, DAP is highly expected to be a proper choice for fire-proof polymers.

Figure 1

Strategy for obtaining fire-safe foams by incorporating DAP additive.

Building on the abovementioned knowledge, in this study, the fire-resistance abilities and thermal stabilities of r-PUF prepared from an oligo-ester-ether-diol, a product of chemical recycling of PET wastes using DEG as a glycolysis reagent and DAP as a phosphorus–nitrogen FR, were comprehensively investigated. Furthermore, the corresponding comparisons and characterizations of the fire resistance and thermal stability performances of r-PUF and c-PUF (PUF derived from commercial polyol) in the presence of phosphorus–nitrogen compounds were first introduced in this investigation. Their physicomechanical properties were also studied. Based on the outcomes of our previous studies, a relatively high loading of TPP or ATH in the r-PUF formulation can be reduced via the utilization of DAP as a phosphorus–nitrogen-containing FR. The results of this study also proved that r-PUF/FRs obtained from PET-DEG glycolytic products not only meet all fire-resistance demands for polymer applications but also deal with reducing the accumulation of waste PET bottles in the environment and improving waste disposal.

Results and Discussion

Structural Morphology of the Prepared Foams

The foam density determined as the ratio of foam weight to its geometrical volume and cell morphology are important parameters that greatly affect foam fire retardancy and physical-mechanical properties. Figure shows the microscopy images of the prepared PUFs. As given in Table , r-PUF exhibited a density of 71.0 kg/m3 and a mean cell size of r-PUF of 360 μm, whereas c-PUF formed with a lower density of 45.6 kg/m3 and a larger cell size of 436 μm, indicating that the r-PUF foam formed with fine cells under similar preparation conditions was denser than c-PUF. The different polyol types used in our study could be a reason for the density difference between the final r-PUF and c-PUF foams. The density of the PUF foams tends to increase by the addition of DAP into PUF and its increased content. It is quite common to find that adding a denser additive to the foam material could increase the foam density. Indeed, the foam density increased considerably from 71 kg/m3 for neat r-PUF to 83 kg/m3 and 106 kg/m3 for the r-PUR/DAP foams at 5 and 15 php DAP contents, respectively. Additionally, a density increase as the DAP content increased was observed for the c-PUF and c-PUF/DAP foams.

Figure 2

Microscopy images of the prepared PUFs with and without DAP addition.

Table 1

DAP Loading, Density, Cell Diameter, Combustion Test Results, and LOI Value of the Prepared r-PUFs and c-PUFs

				combustion test
samplesa	DAP loading (php)b	densityc (kg/m3)	average cell diameterd (μm)	horizontal burning (mm/min)	vertical burning	LOI (%)
r-PUF	0	71.0 ± 2.3	360 ± 44	120	No ratinge	20
r-PUF/DAP5	5	83.0 ± 1.3	401 ± 69	80	No ratinge	-(f)
r-PUF/DAP10	10	86.1 ± 0.64	387 ± 78	29	No rating(e)	-(f)
r-PUF/DAP15	15	106.2 ± 3.8	288 ± 29	20	self-extinguishing (2 s/5 s)	24
c-PUF	0	45.6 ± 0.2	436 ± 43	225	No rating(e)	19
c-PUF/DAP15	15	55.4 ± 1.4	427 ± 63	38	No rating(e)	22
c-PUF/DAP30	30	58.1 ± 0.8	364 ± 58	31	No rating(e)	-(f)

r-PUF and c-PUF are represented for PU foams fabricated from the oligo-diol product of PET glycolysis and the commercial polyol, respectively. Subscripted numbers of 5, 10, 15, and 30 denote DAP loadings incorporated into the foams.

Parts per hundred polyols by weight.

Evaluated according to ASTM D1622.

Evaluated from microscopy images.

The flames combusted up to the specimen-holding clamps.

Not measured.

The presence of DAP at different contents in the reactant mixtures affected not only the density but also the cell structure of the final foams. For all neat PUF and PUF/DAP foams, the mean cell sizes tended to decrease by DAP incorporation, and the DAP content increased. A similar observation in terms of the density and cell size change for foams by the addition of additives was also reported in the literature.[29,30] During the formation and growth of cells, DAP and most additives could play the role of heterogeneous nucleation sites, resulting in a reduction in the foam cell size. For instance, the mean cell diameters were remarkably changed from 360 μm for neat r-PUF to 288 μm for r-PUF/DAP15.

Notably, r-PUF/DAP at 15 php DAP formed an even cell structure with a high foam density of 106.2 kg/m3 and a mean cell size of 288 μm, whereas c-PUF/DAP at the same FR loading showed respective values of 55.4 kg/m3 and 427 μm (Table ). Consequently, these considerable differences in density, mean cell size, and morphology between r-PUFs and c-PUFs were likely to affect the thermal behavior, combustibility, and mechanical properties of the PU foams.

Fire Performance

The two combustion tests of the vertical and horizontal burn methods were applied to determine the flammability of the PUF samples in this study. An extra surface combustion test for the bulk foam materials and the LOI measurement was also used to illustrate the efficiency of DAP in improving the flame resistance of PUF. The test results and detailed descriptions and evaluations of all prepared PUF samples are given in Table , and sample photos taken during the combustion tests are shown in Figures and 4.

Figure 3

Photographs of the c-PUF (a) and r-PUF (b) with and without DAP after 30 s of burning during the horizontal burning test.

Figure 4

Photographs of the c-PUF and r-PUF with and without DAP after 5 s of burning during the vertical burning test.

The PU foams are combustible materials that are severely burned during the tests. The c-PUF and r-PUF exhibited a similar outcome of completed combustion, which was followed by an evaluated failure and combustion at different rates. Indeed, r-PUF burnt at a burning rate of 120 mm/min, which was much slower than the speed at which c-PUF burned, 225 mm/min, indicating the better fire-resistance property of r-PUF. Similar to the horizontal burning test, Figure shows rapid and widespread flame growth for c-PUF, whereas the combustion of r-PUF took place more slowly (Table ). However, no dripping of any PUF specimens was observed during the tests. The LOI measurement evaluating the minimum oxygen percentage needed to maintain combustion for at least 3 min was used to illustrate the combustion test results. Principally, materials with LOI values below 21% are estimated to be highly combustible, whereas higher LOI values mean more oxygen is required to support combustion.[31] The measured LOIs of r-PUF and c-PUF are comparable, with values of 20 and 19%, respectively, demonstrating their high flammability.

The incorporation of DAP into PUFs led to remarkable improvements in reducing the combustibility of the PUF and even achieved a fire self-extinguishing foam at proper FR loadings. Despite no rating being achieved at a small DAP amount of 5 php, r-PUF/DAP5 burnt at a decreased rate of 80 mm/min, and the flame spread more slowly for r-PUF/DAP5 than for neat r-PUF, which had a burning rate of 120 mm/min (Table and Figures and 4). As expected, when more DAP was added into the foam, and better fire resistance was observed. A significantly decreased rate of 29 mm/min was achieved for r-PUF with 10 php of DAP. Raising the DAP loading to 15 php, r-PUF/DAP15 expectedly exhibited the best fire-resistant performance among all PUF/DAP foams prepared in this study. Figures and 4 clearly show that r-PUF/DAP15 specimens were successfully fire-resistant, with combustion stopping 5 s after the removal of the 2nd flame application.

The combustion test with the specimen dimensions of 50 × 50 × 25 mm3 applied to the foam material also presented a similar observation, with no sign of subsequent combustion and no severe damage to the tested specimen (Figure S1). Notably, the neat r-PUF sample was charred completely by its ignition under burning, whereas the specimens of r-PUF/DAP15 endured flame ignition and kept their original shapes, demonstrating the high efficiency of DAP in improving the flammability of r-PUF, as expected. Additionally, with an LOI value of 24%, r-PUF/DAP15 was classified as a self-extinguishing material, which agreed well with its combustion results.

The fire resistance and decomposition mechanism of DAP shares some similarities to other phosphorus–nitrogen intumescent FRs.[32−34] Reportedly, initial DAP decomposition generally produces volatile gases of NH3, H2O, and phosphorus acid. While the generated volatiles can physically serve as dilute combustion fuels and cool down the combustion temperature, the produced phosphoric acid can transform into active phosphorous radicals in a gas phase to deactivate H• and OH• radicals during combustion. In addition, the contribution of the condensed phase via pyrophosphoric acid and other phosphorus-containing residual chars to the flame retardancy of the polymer material was also involved.

The FR performance of c-PUF/DAP improved. The c-PUF samples with DAP loadings of 15 and 30 php all passed the horizontal burning test, which was defined as no burning being sustained after 30 s of flame application. Although the use of the optimized DAP loading of 15 php resulted in the fire-extinguishing r-PUF sample, it was insufficient for c-PUF to meet any rating specification. The vertical specimens of c-PUF/DAP combusted up to the sample-holding clamp and finally remained as a completely charred residual, even with a DAP loading reaching 30 php.

The cone calorimeter is a performance-based fire testing apparatus that correlates well with large-scale fire tests and is commonly used to determine the FR performance of materials.[35] Heat release rate (HRR), peak heat release rate (PHRR), and total heat released (THR) of c-PUF/r-PUF with 15 php DAP loading were recorded and are shown in Figure . Both PHRR and THR of r-PUF/DAP15 were lower in comparison to those of c-PUF/DAP15. The HRR of c-PUF/DAP15 increased rapidly at the beginning of the test and reached the maximum value, PHRR, faster than that of r-PUF/DAP15. Moreover, the PHRR of c-PUF/DAP15 was also higher than that of r-PUF/DAP15 which resulted in a higher and sharper peak than r-PUF/DAP15. The lower PHRR and THR indicated that a higher residual char layer was formed during combustion and increased the FR properties of r-PUF/DAP15. Last, the duration for the combustion to complete for r-PUF/DAP15 was longer when compared to that for c-PUF/DAP15. All above-mentioned results illustrate that the fire-resistance performance of the r-PUF foams derived from the recycled PET products was considerably better than that of c-PUF sourced commercially.

Figure 5

HRRs and THRs of c-PUF/DAP15 and r-PUF/DAP15.

Figure depicts the correlation between the apparent density of PU foams and their fire performance in our study. Despite the principal effect of DAP addition and its loadings in reducing the flammability of PU foam, the fire-resistance performance of c-PUF and r-PUF improved with the apparent density increase. In addition, another reported PU foam, derived from bis(2-hydroxyethyl) terephthalate from postconsumer PET bottles, with a higher apparent density of 114 kg/m3 exhibited better flame resistance for the fire extinguished,[36] further evidencing the correlation.

Figure 6

Correlation between the density and fire performance of PUF and PUF/DAP.

Low density gives rise to quick melting and high flame propagation.[37] The produced melt promotes severe dripping that increases fire hazard. The density value of c-PUF was 45.6 kg/m3, and no ranking in combustion tests was attained; in contrast, the densities of c-PUF/DAP15 and c-PUF/DAP30 were 55.4 and 58.1 kg/m3, respectively, and their flame behaviors were notably improved. Similar results on the density and flame performance of the r-PUF system were obtained by adding 5–15 php of DAP to rPUF. As expected, the density increased gradually (from 71.0 to 83.0–106.2 kg/m3), and the fireproof performance was significantly improved (Figure ).

Higher density means that the fire retardant behavior is better because of a more compact burned layer and less flame spread. It is assumed that the amount of combustible volatiles produced during thermal degradation is proportional to the corrected density of the foams. A study by Biedermann et al. showed that a reduction in the cell size yielded a reduction in heat transfer and total thermal conductivity for PU foams,[38] which are likely to slow down the combustion and flame spread of the PU foams. Additionally, a similar finding of the apparent density effect on the fire-resistance property of PUF modified by expandable graphite or ammonium polyphosphate was reported by Yang et al.[39] Thus, high density leads to a lower flame spread rate.

Thermal Behavior and Fire-resistance Modes

According to the fire performance discussed in the previous section, PUFs without and with 15 php of DAP are selected for Thermogravimetric analysis (TGA) measurements to investigate their thermal endurability. The results, including the onset temperature, the temperature values of T10, T40, and T60 at which the foams lose 10, 40, and 60% in weight, respectively, and the residual content of the foams at 800 °C, are given in Table . The thermogravimetric (TG) and derivative TG (DTG) curves of the foams are shown in Figure .

Table 2

Thermal Properties of c-PUF and r-PUF with and without DAP

samples	Tonset (°C)	T10 (°C)	T40 (°C)	T60 (°C)	char residues at 800 °C (%)
r-PUF	280	304	388	514	24.9
r-PUF/DAP15	262	284	346	687	33.6
c-PUF	240	279	330	354	13.7
c-PUF/DAP15	234	273	324	360	19.7

Figure 7

Experimental TG and DTG curves of neat PUF and PUF/DAP.

The degradation of both r-PUF and c-PUF occurred through two stages: one stage was the temperature range of 200–450 °C, and the other stage was the 450–600 °C range. However, distinct mass-loss rates were observed for the degradation of r-PUF and c-PUF. The first stage with a major weight decrease was attributed to the fragmentation of urethane linkages into isocyanate and polyol segments, and the second step corresponded to the further decomposition of the pyrolysis products and char residues.[40−42] The TG and DTG curves in Figure indicate that r-PUF derived from oligo-diol-containing aromatic compounds was much more thermally stable than c-PUF, as its two DTG peaks were significantly less intense and were observed in higher temperature ranges. c-PUF started to decompose at 240 °C and lost a mass of ca. 70% by the first degradation stage, whereas r-PUF foam was more stable, with a higher Tonset of 280 °C and a mass loss of ca. 50%. Furthermore, T40, T60, and the residue content at 800 °C of r-PUF were considerably higher than those of c-PUF, confirming that r-PUF has better thermal stability.

By comparing the TG and DTG curves of PUFs, the PUF/DAP foams also showed two decomposition stages in the temperature range from 200 to 600 °C overall but with a gradual weight loss at the second step instead. Moreover, it is obvious that Tonset, T10, and T40 of the two PUF/DAP samples were lower than those of neat PUFs because of the decomposition of DAP in the mixtures at the early stage. Indeed, r-PUF/DAP15 exhibited Tonset, T10, and T40 values of 262, 284, and 346 °C, respectively, which were all lower by approximately 18–42 °C than the corresponding temperatures of 280, 304, and 388 °C of r-PUF. On the other hand, the second degradation stage of PUF/DAP shifted to a higher temperature range and had a gradual weight decrease compared to the weight of PUF, which was believed to be a consequence of the early degradation of DAP. The r-PUF/DAP15 foam lost 60% of its weight at 687 °C and retained a charred content of ca. 33.6% at 800 °C, while these values for r-PUF were 514 °C and 24.9%, illustrating the charring acceleration of DAP. The thermal stability of c-PUF also changed by DAP incorporation, confirming the effect of DAP on the thermal behavior of the PU foams. The thermal stability and combustion behavior results obtained for PUFs agree with each other, as discussed in the above section.

Additional experiments in which the PUF and PUF/DAP15 foams were incinerated in a furnace at various temperatures of 300, 500, and 700 °C were carried out to further understand how the foams alter during combustion and degradation. Figures and S2 and Table present the appearance, weight loss trend, and remaining content of the as-prepared and incinerated foams recorded after each temperature, respectively.

Figure 8

Photographs of the as-prepared and incinerated foams at different temperatures.

Table 3

Residues of r-PUFs and c-PUFs with and without DAP after Incineration at Various Temperatures

Temp. (°C)	the residual char (%)
	c-PUF	r-PUF	c-PUF/DAP15	r-PUF/DAP15
300	60.0	69.6	54.4	67.2
500	12.5	20.3	15.8	24.2
700	0.0	1.90	6.80	12.5

A small residue content of 1.9% (Figure , Table ) remained at 700 °C for r-PUF, whereas almost no char of c-PUF remained as incineration was completed. The PUF/DAP15 foams clearly showed char contents of 12.5% for r-PUF/DAP15 and 6.8% for c-PUF/DAP15, which were considerably higher than those for their corresponding PUFs, as expected. Notably, the foams with 15 php of FR were more thermally stable in the high-temperature range and retained better shapes during incineration. The additional incineration outcomes are consistent with the above combustion discussion and TG analysis, confirming the enhancement in the thermal endurance and combustibility of PUFs by DAP addition and indicating the better performances of r-PUFs because of oligo-diol contributing aromatic-containing structures than c-PUFs.

Physical and Mechanical Properties

The physical and mechanical properties, such as the density, morphology, and chemical ingredients, strongly affect the fire retardancy of PUF. A PUF FR with good physical properties can be manufactured by a type of polyol. The r-PUF derived from oligo-diol-containing aromatic compounds was stiff; consequently, the density of r-PUF was high, and r-PUF improved the load-bearing properties. This is why the compressive strength values of r-PUF and r-PUF/DAP were significantly higher than those of c-PUF and c-PUF/DAP. The compressive strengths steadily increase with increasing the DAP content in PUF (Figure ). Cell morphology is also one of the most important factors that affects the physical and mechanical properties of PUFs. No DAP agglomeration was found in the microscopy images, which indicated good compatibility between DAP and PUF. Compared with the cells of the c-PUF system (Figure ), those of r-PUF and r-PUF/DAP were smaller and more uniform in size and exhibited thicker cell struts, which may be responsible for the improvements in the compressive strength.

Figure 9

Compression stress–strain curves of PUF and PUF/DAP.

Filler particles act as nucleation sites for cell formation, and more cells start to nucleate at the same time. There is less gas available for their growth, which leads to a decrease in the size of the cell. Furthermore, the incorporation of unreacted DAP into the PUF causes an increase in the viscosity of the mixture, which inhibits cell growth and results in smaller cell sizes compared to neat PUF.[43,44]

PUFs are reported as being one of the most versatile engineering polymeric foams and are widely used in the construction and building industries. To avoid moisture-related problems in building components and enclosures, understanding moisture storage and transport phenomena is of crucial importance. Hygroscopic materials, such as foam materials, have the ability to adsorb and desorb moisture with variations in surrounding conditions.[45]Figure shows the sorption measurement sequences of the samples under 33, 55, and 75% humidity levels. When the measurement was finished, the equilibrium moisture content by mass (g/g) was measured using eq where m0 and ma are the masses of the sample under the initial dry and controlled relative humidity environments, respectively. The results of the sorption isotherm are shown in Figure .

Figure 10

Sorption measurement sequences of the samples under 33, 55, and 75% humidity levels.

Figure 11

Adsorption and desorption of PUF and PUF/DAP.

The adsorption and desorption isotherm results obtained for PUF/DAP showed almost no significant difference from neat PUF, and both are low moisture capac-ity materials. The sorption isotherms of r-PUF and r-PUF/DAP exhibited normal hysteresis curves; however, the desorption curves of c-PUF and c-PUF/DAP showed higher losses in comparison with the adsorption curves. These losses could be due to the biodegradation of the materials in a high moisture environment. c-PUF was prepared from a commercial polyol that is probably an aliphatic chain that is more easily attacked by fungi and bacteria.[46] Meanwhile, rPUF was derived from PET containing an aromatic ring that is more thermally stable and has less hydroscopic stability. However, these conclusions need further study and discussion.

Conclusions

In general, the outcomes of this study showed that r-PUF with and without DAP showed significantly higher flame retardance, better thermal stabilities, and better mechanical properties than c-PUF with and without DAP. The c-PUF samples with DAP loadings of 15 and 30 php all passed the horizontal burning test, which was defined as no burning being sustained after 30 s of flame application, but the flames combusted up to the specimen-holding clamps in the vertical burning test. Meanwhile, the use of the optimized DAP loading of 15 php resulted in the fire-extinguishing r-PUF. r-PUF/DAP15 and c-PUF/DAP15 had LOI values of 24 and 22%, respectively. The thermal properties of PUF/DAP15 foams were considerably better than those of PUFs, as expected. Notably, the foams with 15 php of DAP were more thermally stable in the high-temperature range and retained better shapes during incineration.

The fire-resistance performance of the r-PUF and c-PUF systems improved with increasing the apparent density. Low density gives rise to quick melting and to high flame propagation. Higher density means that the fire-retardant behavior is better because of a more compact burned layer and a lower flame spread. The compressive strength values of r-PUF and r-PUF/DAP were significantly higher than those of c-PUF and c-PUF/DAP. Compared with the cells of the c-PUF system, those of r-PUF and r-PUF/DAP were smaller and more uniform in size and exhibited thicker cell struts, which may be responsible for the improvements in the compressive strength. The sorption isotherms of r-PUF and r-PUF/DAP exhibited normal hysteresis curves, and the PUF systems were low moisture capacity materials.

Experimental Section

Materials

The diol and polyol reactants used in this study were prepared from two sources of a glycolysis product of wasted PET bottles and a commercial polyol, respectively. Methylene diphenyl diisocyanate (MDI) (Voracor CE101) with 31.0% NCO, viscosity of 210 mPas at 25 °C, and density of 1.23 g/cm3 at 25 °C and commercial polyol (VORACORCR765) were provided by Dow Chemical, Guangzhou, China. The FR DAP was purchased from Guangdong, China. Poly-methyl-alkyl-siloxane, used as a surfactant in PUF preparation, was purchased from BYK, Germany. Catalysts of triethylamine and dibutyltin dilaurate used for PUF preparation were provided by Merck of Germany. All chemicals were used as received without any purification. Distilled water was used as a blowing agent.

Synthesis of Oligo-Diol from Wasted PET Glycolysis

The oligo-diol product is described in detail in our previous reports.[47,48] Briefly, 48.0 g of PET flakes, 66.3 g of DEG, and 0.427 g of zinc sulfate were loaded into an Erlenmeyer flask covered with glass. The PET glycolysis reaction was conducted inside a microwave oven at a constant power of 250 W and a total operation time of 80 min. After a desirable time, the reaction mixture was allowed to cool to ambient temperature, the solid catalyst was removed by centrifugation, and the liquid mixture was obtained and kept in a dry environment to avoid moisture absorption for further applications. In this study, the oligo-diol product was then reacted with a commercial MDI in the presence of a DAP additive to fabricate fire-resistant PUFs.

Preparation of PUFs from Synthesized Oligo-Diol and Commercial Polyol Sources

PUFs were yielded by a reaction of diols and diisocyanates at a fixed ratio in an open mold. In a typical procedure, a mixture of oligo-diol or polyol with and without DAP (5–30 php), the surfactant (2 php), the blowing agent (1 php), and the catalyst (0.1 php) was evenly mixed by a mechanical stirrer at a rate of 700 rpm for 15 min. The mixture was then blended with MDI [oligo-diol or polyol/MDI:1.0/1.3 (g/g)] and stirred vigorously for 15 to 30 s before being transferred into an open mold for growth. A detailed description of the prepared samples is given in Table , where r-PUF and c-PUF denote the foams fabricated from the oligo-diol product of PET glycolysis and the commercial polyol, respectively.

Characterization of the Product

Flammability Tests

The fire-resistance properties were evaluated by the horizontal and vertical burning and LOI tests. For horizontal burning, the specimen was held horizontally at one side, and the other side of the specimen was applied to a burner flame for 30 s. For vertical burning, the burner flame was applied to the specimen for 10 s and then removed until flaming stops at which time the flame was reapplied for another 10 s and then removed. Five specimens with dimensions of 127 × 13 × 10 mm3 were tested for each sample. The burning rate and self-extinguishing time were recorded. The different efficiencies of the flammability behavior of c-PUF and r-PUF with and without DAP could be readily determined by using samples with test bar dimensions of 50 × 50 × 25 mm3 and burning them for 10 s[49] LOI (Qualitest, USA) testing following ASTM D2863 was performed to determine the minimum oxygen concentration at which the materials could ignite to assess their fire resistance. Each sample was prepared with dimensions of 130 × 10 × 10 mm3. The LOI result is the average value of the five as-prepared specimens. The combustion behaviors of PUFs/DAP15 were evaluated by a cone calorimeter (FTT-Fire Technology, East Grinstead, West Sussex, UK) according to ISO 5660-1 under the heat flux of 50 kW/m2. Specimens with dimensions of 100 × 100 × 10 mm3 were wrapped in aluminum foil.

Thermal Stability Analysis

TGA was carried out on a TA instrument (TGA Q500 Universal V4.5A, New Castle, DE, USA) by heating from room temperature to 800 °C at a heating rate of 10 °C/min in a nitrogen atmosphere.

Cell Structure Characterization

The morphology of the foams was determined by using an optical MicroBlue MB.1152 Euromex microscope (Netherlands). The optical images were taken from different areas of each sample, and the average diameter of the cells was calculated according to ASTM D3576-98.

Physical and Mechanical Tests

The apparent density of foam samples with dimensions of 50 × 50 × 25 mm3 was determined as the ratio of the foam weight to its geometrical volume according to ASTM D1622, and the average value of at least three samples was obtained.

The compression test of r-PUF and c-PUF with and without a FR was measured by a universal testing machine (AG-X Plus, Shimadzu, Kyoto, Japan) according to ASTM D1621. The sample with dimensions of 50 × 50 × 25 mm3 was compressed with a loading rate of 2.5 mm/min in the direction parallel to the foam-rising direction.

Sorption isotherm tests of foam samples (50 × 20 × 10 mm3) dried in an oven at 60 °C for 24 h until reaching a constant weight were performed according to the EN ISO 12571 standard. These samples were put in a desiccator at different controlled relative humidity conditions of 33, 55, and 75%, which were prepared with different saturated salt solutions. When the moisture content of all samples reached equilibrium values, the samples were then moved to lower relative humidity conditions for the desorption test. Using this method, the adsorption and desorption isotherm curves were recorded, and the equilibrium moisture content by mass (g/g) was measured.[45,50]