Natural Rubber Latex Wastes from Balloon Production as Valuable Source of Raw Material: Processing, Physico-Mechanical Properties, and Structure

Abstract

This study explores the potential for recycling natural rubber (NR) latex waste from balloon production through the devulcanization and revulcanization processes. The mechanical devulcanization of colored latex balloon waste was conducted, followed by revulcanization using a sulfur-based system. The reclaimed rubber’s properties, including crosslink density, tensile strength, and abrasion resistance, were compared with those of virgin NR. The results demonstrate that the reclaimed rubber maintains a crosslink density close to that of virgin NR. Hardness and abrasion resistance were comparable, indicating successful material recovery. Structural analyses, including FTIR and SEM microscopy, revealed that the devulcanization process effectively allowed for successful revulcanization. This study concludes that NR latex waste can be effectively recycled and reused in rubber composite formulations, offering a sustainable approach to waste management in the rubber industry and contributing to developing eco-friendly materials. In the context of this research, integrating advanced chemical and physical methods, such as solubility parameter calculations and enhanced devulcanization techniques, could further optimize the devulcanization process. These methods quantitatively enhance the efficiency of material recovery, offering a path to more sustainable recycling practices. The findings suggest that combining such advanced methodologies could significantly improve recycled NR latex’s overall performance and applicability in industrial applications.

1. Introduction

In recent decades, there has been a systematic increase in the demand for rubber and natural latex. Global production of natural rubber in 2022 amounted to 14.6 million tons, more than double the production in 2000 when the global production volume was 6.8 million tons [1]. The production of concentrated liquid latex has also systematically increased, reaching over 1.5 million tons (converted to dry rubber) in 2021. The world leader in the production of natural rubber and NR latex is Thailand, accounting for over 37% of global natural rubber production and over 71% of global natural rubber latex production [2].

The increasing consumption affects the growing problem of generated waste. There is increased scientific research accompanying the growing issue of generated waste and waste rubbers management and recycling. The number of published articles in this field has significantly increased in the last two decades [3]. The majority of the research focuses on the recycling of waste tires. These studies have a tangible effect, as reclaimed rubbers are available on the market, and their use in appropriate amounts reduces production costs without significantly lowering the quality of rubber products. Waste tires could find application in the construction industry, using waste rubber granulates and powders as components in asphalt or concrete [4,5]. However, it should be noted that tires are made from a mixture of natural and synthetic rubber. As a result, there is a gap in research concerning less vulcanized rubber waste produced from natural rubber and natural rubber latex.

Due to relatively high rubber content, latex sourcing from rubber trees (Hevea brasiliensis) remains the most profitable. Raw latex can be processed into dry rubber after coagulation, pressing, and drying [6]. Natural rubber (NR) is divided into many categories depending on its origin, quality, and specifications. The dominant categories are RSS (Ribbed Smoked Sheet) and TSR (Technically Specified Rubber), which have narrower specifications. TRS, RSS, and blends of NR account for almost 90% of global production. The primary consumer of NR is the tire industry, which consumes about 76% of natural rubber and 65% of the world’s total rubber production (both natural and synthetic) [7]. A separate process involves processing raw latex into concentrated natural latex.

Concerning the production of concentrated natural latex, the raw latex obtained from rubber tree plantations undergoes processes that increase the dry rubber content to a minimum of 60%. The latex is subjected to clarification and maturation, which significantly increases the colloidal stability of the solution [8].

From a chemical structure perspective, latex is a colloidal solution of cis-1,4-polyisoprene. The polymer chain has α and ω terminal groups linked to phospholipids and proteins, forming a membrane surrounding the spherical rubber particles. The nature of rubber particles is a widely described topic in the literature [9,10], and the colloidal properties of natural latex arise from the unique structure of rubber particles suspended in an aqueous serum. The surface of the spherical rubber particles surrounded by a phospholipid membrane has a negative charge, and additional stabilization in the aqueous serum is provided by naturally forming or added soaps of higher fatty acids [11]. The colloidal stability of latex is also influenced by the alkaline nature of latex’s aqueous serum.

In its unvulcanized form, natural rubber has weaker properties and lower thermal and mechanical resistance. Rubber products are created by crosslinking the rubber in the vulcanization process. Creating crosslinks between polymer molecules leads to shape durability, better elasticity, resilience, tensile strength, hardness, and resistance to weather conditions. A common crosslinking agent is sulfur, which forms sulfide bonds between individual polymer chains in the presence of chemical accelerators and activators [12]. The specific mechanical properties of a material depend on its crosslink density. In other words, the material’s final characteristics are determined by both the quantity and the type of bonds formed. In the case of sulfur vulcanization, the bonds can be mono-, di-, or polysulfidic [13]. The appropriate ratio of accelerators to sulfur influences the formation of different types of bonds. To a certain extent, a higher number of accelerators increases the number of mono-sulfidic bonds relative to polysulfidic bonds. A higher number of monosulfidic bonds increases the stress formation rate due to the rubber sample’s deformation.

Among natural latex recipients, manufacturers of dipped products such as gloves, condoms, or balloons dominate. Most liquid latex is used to produce medical or protective disposable gloves, for which demand increased notably during the COVID-19 pandemic. It is estimated that in 2020, during the pandemic, 65 billion gloves (both natural and synthetic latex) were used monthly [14].

Concerning NR-dipped latex products such as gloves, condoms, or balloons, it can be assumed that the vulcanization process of rubber in liquid form corresponds to the vulcanization process of rubber made from dry rubber. In the case of latex, a liquid emulsion of natural rubber, the commonly used term is prevulcanization, which is the vulcanization of the polymer while minimizing chemical shock, is used to maintain the colloidal nature of the latex solution and preserve its further processing potential. A separate issue is the proper preparation of ingredients added as aqueous dispersions. The degree of dispersion of the ingredients determines their activity and the rate of crosslinking reactions [15]. Prevulcanized latex (PV latex) is a ready-to-use material, and the dried latex film will have the desired mechanical properties.

Dipped products are the most commonly manufactured on the production lines. This process can be simplified using the example of latex balloon production. The molds are dipped in a coagulant solution, followed by PV latex. The coagulant causes the latex to gel evenly on the surface of the molds. Next, the molds with the gelled wet latex are leached in water. Leaching removes the remaining coagulant ions and water-soluble substances from the gel, enhancing the product’s chemical safety and strength. The leached molds are then placed in the anti-tacking agent; next, they go through the oven, where the water present in the gel evaporates. Finally, dry balloons can be stripped from the mold.

Concerning dipped rubber products made from natural rubber latex, despite the raw material used for production being a natural polymer derived from latex trees, products such as balloons have been classified as single-use plastic. They are subject to the restrictions contained in Directive (EU) 2019/904 of the European Parliament and of the Council of 5 June 2019 on the reduction of the impact of certain plastic products on the environment [16]. The reasoning behind this classification is that the biopolymer is processed through chemical crosslinking. Rubber, as vulcanized rubber, has increased durability, which includes, among other things, limited susceptibility to biodegradation. The durability of vulcanized rubber is more outstanding the higher the crosslink density.

There are different approaches to rubber waste. Landfilling waste (even sorted waste) is not a satisfactory solution. Recycling is widely discussed, particularly concerning waste tires. Simultaneously, waste in the form of used tires constitutes up to 80% of rubber waste. One of the seemingly most straightforward possibilities is energy recovery. The high heat of combustion of tire rubber makes it a calorific fuel, for example, in cement plants. However, this process requires installations that provide appropriate combustion conditions to minimize the negative impact on the environment. An alternative energy recovery method could be pyrolysis, during which, in a properly controlled thermal reaction, liquid fuels are obtained from the polymer waste [17].

Recycling of rubber waste can also based on so-called material recovery. In this case, the polymer undergoes devulcanization or reclaiming. The term “devulcanization” can have a broad meaning concerning commonly understood rubber recycling, but in the most precise sense, it refers to processing vulcanized polymer to reduce the number of crosslinks and significantly decrease crosslink density. This task involves selectively breaking sulfur-sulfur (S-S) or carbon-sulfur (C-S) bonds as much as possible without disrupting the main polymer chain, i.e., carbon-carbon (C-C) bonds. It is a complex process because the energies required to break S-S and C-S bonds (227 and 273 kJ/mol, respectively) are not significantly different from the energy needed to break C-C bonds (348 kJ/mol) [18,19]. The more the main chain degrades, the less justified the term “devulcanization” becomes, and the process is then referred as “reclaiming”. Ideally, this process restores the processing capabilities of the devulcanizate, allowing it to be crosslinked again, i.e., revulcanized. Depending on the type of energy used during the process, there can be chemical, thermal, mechanical, and biological devulcanization and numerous modifications thereof [20].

Publications on the recycling of rubber waste made from NR, not mixed with synthetic rubbers, emphasize the high potential of this type of waste. A relatively comprehensive description of the issue of NR latex product waste can be found in Rajan’s studies [21,22]. The author reviewed the most popular forms of devulcanization and a historical overview of material recycling. The research included investigating the thermomechanical devulcanization process of waste material from latex gloves and condoms. The study focused on the impact of reclaiming agents used in the devulcanization process. During subsequent experiments, diphenyldisulfide, hexadecylamine, and 2,2’-dibenzamidodiphenyldisulfide were used. The author conducted a study to compare amine and disulfide as reclaiming agents, demonstrating the greater effectiveness of disulfide. Research on the breakdown of network bonds showed that some polysulfide bonds remain when hexadecylamine is used, compared to the absence of polysulfide bonds in the regenerated material treated with diphenyl disulfide. Another waste reclaiming process was carried out using 2,2’-dibenzamidodiphenyldisulfide, yielding results similar to those obtained with diphenyldisulfide. Both of these compounds were used in a truck tire tread mix, replacing the fresh natural rubber with reclaimed rubber.

Thaicharoen et al. [23] conducted mechanical devulcanization of NR material using thiosalicylic acid and diphenyldisulfide as devulcanizing agents. The results obtained from both substances were comparable. The degree of devulcanization was determined using sol-gel fractions of the devulcanized rubber.

Rooj et al. [24] performed devulcanization of NR using benzoyl peroxide as a devulcanizing agent at varying concentrations. The devulcanization process was carried out using both chemical and mechano-chemical methods. In the chemical devulcanization, the material was treated with a solution of xylene and the devulcanizing agent, with the process conducted for 2–6 h at a temperature of 80 °C. The temperature for mechano-chemical devulcanization was also 80 °C. When comparing the degree of devulcanization and the results of mechanical strength tests, chemical devulcanization proved less effective and required a higher concentration of the devulcanizing agent. The devulcanized material was then mixed with raw rubber and revulcanized. A significant decrease in tensile strength was observed when the concentration of the devulcanizate exceeded 60 phr.

In this study, natural rubber latex wastes from balloon production were treated by mechanical devulcanization using a two-roll mill, which is a good approach for preliminary investigation before the application of continuous thermo-mechanical methods. The processing, physico-mechanical properties, and structure were investigated using a Mooney viscosity test, rubber process analyzer, swelling test, tensile tests, and hardness, resilience, and abrasion resistance measurements.

2. Materials and Methods

2.1. Materials

Natural rubber latex wastes, in the form of latex balloons, were sourced from a balloon factory in the European Union. The post-production rubber waste was sorted by color into clear, white, black, and red. The balloon production process described in the introduction directly relates to the material under investigation.

Natural rubber (NR) (reference material) was obtained from concentrated natural latex (60% DRC) LATZ Centex, sourced from Corrie Maccoll, Zaandam, The Netherlands. The unvulcanized natural rubber samples were prepared in the laboratory using a method based on the balloon production process. Balloon molds were dipped in a coagulant solution of calcium nitrate (Yara, Oslo, Norway), and then in the latex. The gel samples were leached in water for 20 min and dried in a laboratory dryer at 80 °C for 1 h.

In the studies, the following chemicals were used during vulcanization: stearic acid, zinc oxide, TBzTD (tetrabenzylthiuram disulfide), MBT (2-mercaptobenzothiazole), and sulfur supplied by P.P.H. Standard Sp. z o.o. (Lublin, Poland). Ethyl alcohol was purchased from Destylacje Polskie Sp. z o.o. (Oborniki, Poland). Acetone and toluene (analytical grade, Chempur, Piekary Śląskie, Poland) were used in the investigation of vulcanized samples.

2.2. Sample Preparation

2.2.1. Waste Latex Balloons Devulcanization and NR Preparation

Balloon waste in the amount of 500 g of each color was subjected to mechanical devulcanization on a laboratory two-roll mill with a working space of 200 × 400 mm (type 14201/P2, Buzuluk, Komárov, Czech Republic). Friction was 1.08 and the gap width varied between 0.2 and 8 mm. The mastication time was approximately 35 min. During mixing, the rolls were cooled with water. The unvulcanized natural rubber (NR-gelled on balloon molds), subjected to only three passes on the two-roll mill to obtain a homogeneous sheet, was used as a reference sample. The obtained products were coded according to the color of used rubber wastes as D-Clear, D-White, D-Black, and D-Red. The appearance of samples before and after treatment is presented in Figure 1.

2.2.2. Revulcanization of Devulcanized Waste Latex Balloons

Mixtures of the obtained devulcanizates with a sulfur-based curing system were prepared in a Brabender^® internal mixer (type GMF 106/2) (Brabender GmbH & Co. KG, Duisburg, Germany) at a mixing temperature of 60 °C, at 60 rpm, with a total mixing time of 8 min. The samples were vulcanized on a press at 150 °C and a pressure of 100 bar, based on previously determined optimal vulcanization time (t₉₀). Vulcanized samples were prepared for further testing: 2 mm thick samples for cutting dumbbells, discs for rebound elasticity testing (1.5 × t₉₀), and discs for abrasion testing (1.2 × t₉₀).

Table 1. The formulation used for the revulcanization of devulcanized samples.

Ingredient	Amount (phr)
D-Clear/D-White/D-Black/D-Red/NR	100
Stearic acid	2.0
ZnO	4.0
TBzTD	1.6
MBT	0.5
Sulfur	0.6

presents the formulation of the compound used to vulcanize the devulcanized waste latex balloons. The low sulfur content (0.6 phr) reflects the low sulfur content used during the prevulcanization of latex for dipped products. Faster accelerators, such as the one used in this study, are commonly applied during the prevulcanization of latex. The combination of thiuram (TBzTD) and thiazole (MBT) positively influences the stability of the prevulcanization process. TBzTD is especially valued as a replacement for the phased-out TMTD, as it enhances the chemical safety of the vulcanizate, mainly by reducing the formation of harmful N-nitrosamines [25].

2.3. Methodology

The processing properties of the devulcanized waste latex balloons were measured using a Mooney viscometer MV2000 (Alpha Technologies, Wilmington, DE, USA) at 100 °C, based on the ISO 289-1 [26] method.

Minimal torque M_L (dNm), maximal torque M_H (dNm), scorch time (t_s1) (min), and optimal cure time (t₉₀) (min) with a sulfur-based curing system were tested using an Alpha Technologies Premier RPA rheometer (Hudson, OH, USA), according to ISO 6502 [27].

Tensile properties tests were conducted based on the ISO 37 [28] on a Shimadzu EZ-Test machine with an extensometer DSES-1000 (Kioto, Japan). Samples were prepared using a type D dumbbell die, according to ASTM D412 [29].

Hardness measurement was performed based on the ISO 7619-1 [30] norm. A digital Shore hardness tester, type A HPE from Zwick Roel (Ulm, Germany), was used to measure the hardness of the vulcanizates.

The rebound elasticity (rebound resilience) was determined using the Schob type rebound pendulum, according to ISO 4662 [31] with the Gibitre Instruments Srl apparatus, Bergamo, Italy.

The abrasion resistance was measured using the Schopper-Schlobach method, according to ISO 4649 [32], using a rotating cylindrical drum device from Gibitre Instruments Srl, Bergamo, Italy. The samples were weighed before testing on a microbalance (AS 220 C2 Radwag, Radom, Poland), and then abraded with 60 grit paper. The pressure on the sample was 10 N. During the test, the sample travelled a total friction distance of 40 m while rotating around its axis. The abrasion resistance was calculated according to Equation (1):

$$\Delta V_r={\displaystyle \frac{\Delta m_t}{{\rho }_t}}\times {\displaystyle \frac{0.2}{\Delta m_{ref}}}\times 1000$$

(1)

where ΔV_r is the abrasion resistance (mm³), Δm_t is the mass loss of the tested sample (g), Δm_ref is the defined value of the mass loss for the reference compound (g), and ρ_t is the density of the sample (g/cm³).

The density of the vulcanizates was determined using the Archimedes method. Samples were weighed in air and ethyl alcohol using a Radwag AS 220.R2 PLUS scale (Radom, Poland), equipped with a density measurement kit KIT 85.

Based on data from the literature, the crosslink density was determined using the equilibrium swelling method [33]. Revulcanized samples weighing approximately 0.2 g were initially rinsed in acetone for about 48 h. After this time, the samples were dried at 60 °C until a stable mass was obtained. Subsequently, the samples were placed in toluene for 72 h. Closed containers with the samples in solvents were kept in the dark at room temperature (24 ± 1 °C). Using a weighing container, the mass of swollen samples was determined (Radwag AS 220.R2 PLUS). Then, the samples were dried at 70 °C, and their mass after drying was determined.

The crosslink density and average molecular mass between crosslinks were calculated, based on the Flory–Rehner Equation (2):

$${\nu }_c={\displaystyle \frac{\ln (1-V_r)+V_r+\chi V_r^2}{2{\rho }_r{V}_s\left(\sqrt[3]{V_r}-\frac{V_r}{2}\right)}}={\displaystyle \frac{1}{{2M}_c}}$$

(2)

where ν_c is crosslink density (mol/g), V_r is volume of the polymer in the swollen sample (cm³)—Equation (3), V_s is molar volume of the solvent (cm³/mol) (for toluene it is 106.27 cm³/mol), M_c is average molecular weight between polymer crosslinks, and χ— is the Flory-Huggins polymer-solvent interaction parameter for the NR—toluene system = 0.414 based on literature data [34].

The volume of rubber in the swollen sample (V_r) was calculated according to Equation (3):

$$V_r={\displaystyle \frac{\frac{m_{dry}}{{\rho }_r}}{\frac{m_{dry}}{{\rho }_r}+\frac{m_{sw}-m_{dry}}{{\rho }_s}}}$$

(3)

where m_dry is the mass of the sample after evaporation of toluene (g), m_sw is the mass of the swollen sample (g), ρ_r is the density of the rubber sample (g/cm³), ρ_s is the density of the solvent (g/cm³), and the density of toluene is 0.867 g/cm³.

The degree of swelling of the sample Q_r was calculated according to Equation (4):

$$Q_r={\displaystyle \frac{m_{sw}-m_1}{m_1}}\times 100\%$$

(4)

where m_sw is the mass of the swollen sample (g) and m₁ is the mass before placing it in toluene (g).

The difference in mass of the sample before and after rinsing with acetone was used to determine the percentage content of non-rubber substances leached out by acetone according to Equation (5):

$$Acetone extract={\displaystyle \frac{m_0-m_1}{m_0}}\times 100\%$$

(5)

where m₀ (g) is the initial mass of the sample prepared for testing and m₁ (g) is the mass of the sample after rinsing with acetone.

The difference in mass between the sample before immersion in toluene and the dry sample after evaporation of toluene was used to determine the sol fraction of the sample according to Equation (6):

$$Sol fraction={\displaystyle \frac{m_1-m_{dry}}{m_1}}\times 100$$

(6)

where m₁ (g) is the mass of the sample before immersion in toluene and m₂ (g) is the mass of the sample after drying from toluene.

The Devulcanization degree D% was calculated according to Equation (7):

$$Devulcanization degree={\displaystyle \frac{{\nu }_2-{\nu }_1}{{\nu }_1}}\times 100$$

(7)

where ν₁ and ν₂ are the crosslink density of samples before and after devulcanization.

The chemical structure was examined using Fourier-transform infrared (FTIR) spectroscopy. The measurements were performed using an IRTracer-100 spectrophotometer manufactured by Shimadzu (Kioto, Japan). A diamond prism objective was used for the measurements. Spectra were obtained in the range of 4000–400 cm⁻¹.

The morphology of the surfaces created by breaking the samples in the tensile test at the speed of 500 mm/min was evaluated using a FlexSEM 1000 II scanning electron microscope (Hitachi, Tokyo, Japan). During the analysis, the electron beam accelerating voltage was 10 kV. Before measurement, the samples were coated with a fine gold layer to increase their conductivity in the vacuum chamber, using a 108 Auto Sputter Coater from Cressington (Watford, UK).

3. Results and Discussion

3.1. Mooney Viscosity

It should be noted that in the production of dipped rubber products, the reason for rejecting a product as non-compliant is often not due to the lower quality of the rubber itself. For example, the product (balloons or gloves) is rejected as non-compliant due to air inclusions. As a result, this product loses its usability and is rated as a waste, but NR matrix retains the same quality in terms of material. Sorting the waste by color was intended to help with a more precise characterization of such production waste.

Table 2. Composition of waste balloons and Mooney viscosity of devulcanizates samples.

Waste Balloons Composition	Devulcanizate Sample Coding	Mooney Viscosity ML (1 + 4) 100 °C
Clear balloons—vulcanized NR material without additional pigmentation	D-Clear	50.6
White balloons—vulcanized NR material pigmented by TiO2 (<9%)	D-White	48.2
Black balloons—vulcanized NR material pigmented by carbon black (<3%)	D-Black	59.0
Red balloons—vulcanized NR material pigmented by less amount of TiO2 (<3%) and organic pigments (<1%)	D-Red	60.9

presents the characteristics of the individual samples with a particular focus on the presence of TiO₂ and carbon black.

The content of these additives is significantly lower than that of standard NR vulcanizates or industrial mixtures. The highest content of non-rubber components was found in the white balloon waste samples (B-White), which contained less than 9% titanium dioxide (TiO₂). The black balloon waste (B-Black) contained carbon black, the content of which did not exceed 3%. The TiO₂ content in the red balloon waste (B-Red) did not exceed 3%, and the content of the red dye was below 1%. The exact content of additives cannot be disclosed due to the balloon manufacturer’s technological secrecy.

The long time (35 min) required to obtain a homogeneous sheet of devulcanized material was partially due to the large amount of waste relative to the working space of the roller mill. It should be noted that no additional devulcanization agents were used at this stage. In each case, the material temperature increased to around 50 °C during mechanical processing on the two-roll mill, despite cooling the rollers with cold water. The self-heating of the material indicates the action of shear forces, which are primarily responsible for the mechanical devulcanization of tested samples. It can be observed, here, that during mechanical devulcanization, the dominance of shear forces can benefit at lower system temperatures (80–100 °C), similar to the case studied. This self-heating phenomenon is sometimes explained by a better dispersion of shear forces compared to thermal devulcanization when externally heating the system [35,36]. Table 2 shows the Mooney viscosity of the obtained reclaimed rubbers.

The Mooney viscosity of studied materials ranged from 48.2 MU to 60.9 MU. A decrease in the viscosity of the reclaimed rubber may indicate a higher degree of devulcanization or polymer degradation. At the same time, there is no clear relationship between the content of coloring additives and the increase in the viscosity of the devulcanized waste latex balloons, as the devulcanized waste latex balloons with the highest content of additives (D-White) had the lowest viscosity (48.2 MU). Furthermore, reclaimed rubber with carbon black had a similar viscosity to the sample D-Red.

Unvulcanized NR obtained from dried latex as a reference rubber had too high molecular weight to be measured with the Mooney viscometer (>200 MU for initial value). The high viscosity of the used NR may indicate a higher average molecular mass of cis-1,4-polyisoprene obtained from concentrated latex (LATZ; DRC 60%) than the standardized dry NR production process (in blocks).

3.2. Curing Characteristics

Figure 2 presents the vulcametric curves of the prepared compounds, while

Table 3. Curing parameters determined for reclaimed rubbers with curing system.

Parameter	R-Clear	R-White	R-Black	R-Red	V-NR
Minimal torque ML (dNm)	0.5	0.5	0.7	0.8	0.9
Maximal torque MH (dNm)	6.7	7.4	7.5	7.2	5.4
Torque increment ΔM (MH-ML) (dNm)	6.2	6.9	6.9	6.4	4.5
Scorch time (ts1) (min)	1.8	1.7	1.3	1.4	2.6
Optimal cure time (t90) (min)	8.2	7.1	5.5	5.5	10.7

shows the key parameters obtained for the test vulcanization. The curves did not exhibit reversion characteristics, reaching a plateau over the 30-minute test duration. It is a favorable result, as it indicates the absence of network degradation or devulcanization during prolonged vulcanization. The data reveal differences between the reference sample (V-NR), and the revulcanized samples.

The minimum torque values (M_L) of the devulcanizates were lower than the M_L of the reference NR sample (V-NR). This result corresponds to the previously measured Mooney viscosity values, where higher Mooney viscosity was associated with higher M_L values during vulcanization. This parameter relates to the processability of the compound and its flow behavior. Therefore, the reclaimed rubber (R) samples could have achieved better results due to easier mixing than the reference sample (V-NR). On the other hand, the maximum torque value (M_H) relates to the hardness of the vulcanized samples. The revulcanizates R-White and R-Black, containing TiO₂ and carbon black present in white and black balloon waste, exhibited the highest M_H values (7.5 dNm and 7.4 dNm).

The torque increase ΔM (M_H-M_L) reflects the obtained crosslink density of the revulcanizate samples and the reference vulcanizate V-NR. In the case of revulcanizates, higher ΔM values may result from the presence of additional curing agents present in the reclaimed rubbers (D-Clear, D-White, D-Black, and D-Red). For example, a similar effect was observed in the test revulcanizates of devulcanized ground tire rubber [37]; the authors suggest that the recombination of broken bonds may occur during devulcanization, making the devulcanization incomplete. Simultaneously, the devulcanizate may contain sulfide half-bridges, increasing the final crosslink density at higher temperatures during revulcanization and influencing the optimal vulcanization time (t₉₀). Lewin et al. [38] also suggest that sulfide residues bonded with the polymer in the devulcanizate are responsible for higher crosslink density during revulcanization. The vulcanization time (t₉₀) of the reclaimed rubbers obtained from balloons was shorter than the t₉₀ of the reference sample, which may confirm these theses.

3.3. Acetone Extract and Swelling Behavior

The results of density measurements and crosslink density are presented in

Table 4. Density and swelling tests results performed for the samples of the wastes (B) obtained devulcanizates (D) and the revulcanizates (R); (Flory–Huggins polymer-solvent interaction parameter: χ = 0.414; the molar volume of toluene: V_s = 106.27 cm³/mol; and the density of toluene ρ_s = 0.867 g/cm³.

Property	Clear			White			Black			Red			V-NR
	B	D	R	B	D	R	B	D	R	B	D	R
Density ρr (g/cm3)	0.935	0.935	0.935	0.994	0.994	0.994	0.949	0.949	0.949	0.956	0.956	0.956	0.931
Acetone extract Fac (%)	2.4	0.8	3.3	2.0	1.6	2.9	2.5	1.3	2.9	2.7	1.3	3.0	3.2
Sol fraction Fsol (%)	0.7	16.8	1.9	0.7	20.0	1.9	0.5	18.6	1.9	1.0	13.6	2.1	1.4
Swelling index Qr (%)	552	1264	403	502	1107	351	535	1114	357	525	1005	382	456
Crosslink density νc (×10−5 mol/g)	3.18	0.48	5.84	3.19	0.48	6.06	3.24	0.55	6.74	3.25	0.74	5.78	4.54
Average molecular weight Mc (×103)	15.72	103.21	8.55	15.67	104.53	8.25	15.45	89.66	7.42	15.38	67.69	8.65	11.01
Devulcanization degree D% (%)	-	84.7	-	-	85.0	-	-	82.8	-	-	77.3	-	-

.

An evident influence of non-rubber substances is observed in density measurements, especially for balloons containing TiO₂ with a high specific gravity (4.26 g/cm³). Other waste samples containing less TiO₂, carbon black, or dyes with lower density than TiO₂ will have lower density. In this context, the full range of balloon waste density from 0.935 to 0.995 g/cm³ is evident.

The fraction extracted by acetone (F_ac) refers to removing residues of crosslinking agents, antioxidants, and some colorants from the sample. The value of F_ac for the revulcanized samples averaged 3.0% and did not differ significantly between the revulcanizates and the reference sample (V-NR). The F_ac content of balloon samples was slightly lower than that of revulcanizates samples, averaging 2.4%, with no significant difference within their color group. Partial leaching occurred with the pigments from the B-Red, D-Red, and R-Red samples, coloring the solvent red. The colorants in balloons are organic, making them more easily leached by organic solvents. They were also not as firmly bound to the polymer as TiO2 or carbon black, which were not leached by solvents. Concerning the balloon waste samples (B), acetone washing removed the anti-tack agent and powder on the latex film’s external and internal surfaces. Lower acetone extract content in the case of the devulcanizates (D) may be explained by the evaporation of additives or pyrolyzed products resulting from the degradation of the network during mastication.

The sol fraction (F_sol) indicates, among other things, the amount of non-crosslinked polymer leached from the sample swollen in toluene. It can be noted that the relatively high purity of the material in the sample, the relatively low content of additives, and the absence of fillers in higher amounts, may have led to some non-crosslinked polymers remaining in the swollen sample—a possible reason for the low sol fraction. The highest F_sol was recorded for the devulcanizate samples (D), ranging from 13.6% to 20.0%. The observed increase in the sol fraction is a sign of devulcanization, and its correlation with the devulcanization degree D% is evident. The content of F_sol found in the revulcanizate samples (R, an average of 2%) was higher than the reference sample (V-NR 1.4%). The F_sol content of the revulcanizates was slightly higher than that of the balloon samples, which averaged 0.7%. This difference may be a typical effect of mechanical treatment, particularly since the devulcanization of the samples was carried out without additives that generally enhance the selectivity of the process, potentially leading to more significant degradation of the main chain.

The samples’ swelling index (Q_r) is related to the crosslink density. Significant swelling of the devulcanized samples (D) was observed. At the same time, such swollen samples maintained their integrity, indicating, on the one hand, incomplete devulcanization and, on the other hand, a high molecular mass of the chains between the remaining crosslinks. It should be noted that the swelling index would be a suitable and easy-to-calculate parameter when comparing samples of the same density. In the case of the tested samples, whose specific gravities varied due to differences in composition, it seems more appropriate to calculate the crosslink density expressed as ν_c (mol/g).

The crosslink density of the samples before and after devulcanization was used to calculate the devulcanization degree. The least devulcanized sample was D-Red, while the most devulcanized was D-White and D-Clear—these were the samples with the lowest viscosity. No significant correlation was observed between the degree of devulcanization and the type of coloring agent present in the various waste series.

The revulcanized samples obtained higher crosslink densities than the virgin V-NR sample, likely due to the presence of residual curing chemistry from balloons and better dispersion of the curatives due to lower viscosity of polymer matrix. This effect was evident on the vulcanization curves discussed earlier (see Figure 2 and Table 3). The ν_c values of the non-reclaimed samples (B) were significantly lower and showed a much narrower range. The differences in the ν_c values of the revulcanized samples may indicate a high susceptibility of the material to degradation during devulcanization. However, the highest ν_c value of the R-Black sample could partially result from the carbon black acting as an active reinforcement filler in the polymer matrix.

3.4. Mechanical Properties

The tensile strength and hardness results of the revulcanizates were compared with the tensile strength test results of the PV latex used for the production of balloons, which constituted the separated waste samples B-Clear, B-White, B-Black, and B-Red. The dumbbells were cut from the flat parts of large balloons for these samples. These samples’ thickness was only 0.39 ± 0.02 mm. This thickness is lower than that specified in ISO 37 but closely relates to the thickness of the waste material studied in this work. Figure 3 presents the elongation curves of the revulcanizates and latex films. The material before devulcanization, and additionally, the results of tensile properties are presented in

Table 5. Tensile properties of revulcanizates (R-Clear, R-White, R-Black, R-Red), reference sample V-NR, and samples of thinner films of PV latex (B-Clear, B-White, B-Black, and R-Red).

Property	Samples of Vulcanizates				Samples of PV Latex Films
	R-Clear	R-White	R-Black	R-Red	V-NR	B-Clear	B-White	B-Black	B-Red
Tensile strength (MPa)	20.6 ± 1.4	21.6 ± 1.2	19.1 ± 1.7	21.6 ± 1.4	19.4 ± 1.9	27.6 ± 1.4	28.7 ± 0.8	25.4 ± 1.1	25.8 ± 2.0
Elongation at break (%)	658 ± 9	604 ± 11	608 ± 8	666 ± 12	705 ± 27	837 ± 14	848 ± 14	810 ± 14	807 ± 13
Modulus at 100% (MPa)	0.85 ± 0.01	0.97 ± 0.01	0.96 ± 0.01	0.87 ± 0.01	0.78 ± 0.01	0.70 ± 0.01	0.71 ± 0.01	0.69 ± 0.01	0.73 ± 0.01
Modulus at 300% (MPa)	1.9 ± 0.0	2.5 ± 0.0	2.4 ± 0.1	2.1 ± 0.0	1.6 ± 0.0	1.1 ± 0.1	1.3 ± 0.0	1.2 ± 0.0	1.3 ± 0.0
Modulus at 500% (MPa)	5.4 ± 0.1	9.8 ± 0.2	8.2 ± 0.7	6.8 ± 0.3	3.6 ± 0.3	3.4 ± 0.2	4.1 ± 0.1	4.3 ± 0.2	4.1 ± 0.2

.

There is a noticeable difference between the results obtained for PV films and press-vulcanized samples. Despite the lower crosslink density, the PV latex film samples B-Clear, B-White, B-Black, and B-Red had the highest tensile strength and highest elongation values. Also in this case, the effect of TiO₂ on the increase in tensile strength was noticeable. The PV latex samples were vulcanized in the liquid latex stage, which resulted in better curative dispersion. The material was produced by the dipping method, following controlled coagulation, washing, and drying. All of these steps result in better coherence and uniformity in the structure of rubber film.

In the case of the samples vulcanized in the press (R and V-NR), the obtained tensile strength results were similar to those of the reference sample V-NR. The most comparable sample, R-Clear (without coloring additives), had even higher tensile strength than the V-NR sample. The R-White and R-Red samples exhibited higher tensile strength than R-Clear, which could be an additional effect of TiO₂ and pigments. However, a similar impact of carbon black was not observed in the R-Black samples. Comparing the tensile strength of the R and V-NR samples, no significant polymer degradation due to devulcanization was observed. The noticeable difference in elongation at break may be due to differences in obtained crosslink density and partly due to coloring additives (TiO₂ in the R-White). Crosslink density is inversely proportional to the average molecular mass of the chain between crosslinks. Longer chains allow for greater freedom in stretching the sample.

The modulus 100 and 300 values also correlate with the measured crosslink density values. For highly stretchable elastomers, the tensile that appears in the initial stretching stage can be used to compare the hardness of the tested samples. The correlation between the crosslink density and M300 is presented on Figure 4a and correlation between crosslink densities and elongation at break is presented on Figure 4b.

Table 6. Mechanical properties of revulcanizates (R-Clear, R-White, R-Black, R-Red), reference sample V-NR, and samples of PV latex films (B-Clear, B-White, B-Black, and R-Red).

Property	Samples of Vulcanizates				Samples of PV Latex Films
	R-Clear	R-White	R-Black	R-Red	V-NR	B-Clear	B-White	B-Black	B-Red
Hardness (Shore A)	32.7 ± 0.5	34.7 ± 0.4	36.0 ± 0.5	33.5 ± 0.4	31.4 ± 0.7	30.9 ± 0.5	31.8 ± 0.8	34.7 ± 0.7	33.0 ± 0.4
Abrasion resistance (mm3)	176 ± 15	196 ± 10	161 ± 15	174 ± 12	150 ± 9	-	-	-	-
Rebound resilience (%)	75.9 ± 1.0	76.9 ± 0.7	74.5 ± 0.8	74.7 ± 0.9	77.2 ± 0.8	-	-	-	-

presents the measurement results of the remaining mechanical properties, hardness, resilience, and abrasion. Additionally, it was possible to determine the hardness of balloon waste material samples (Shore A).

The tested samples of the revulcanizates were characterized by relatively low hardness. The slight differences between the individual samples were partially due to the content of coloring additives. Moreover, the hardness of the samples showed some correlation with the crosslinkdensity and the tensile measured during strength tests, as shown in Figure 5.

The abrasion values of the reclaimed rubbers ranged from 161 to 196 mm³. All devulcanizates had slightly higher abrasion values than the reference sample V-NR (150 mm³). It can be a typical effect of the deterioration in material quality resulting from recycling. The obtained abrasion results might also have been more significantly influenced by the content of coloring additives. Among the revulcanizates (R), it was noted that the sample R-White, with the highest TiO₂ content, exhibited the highest abrasion. While the presence of TiO₂ increased abrasion, the presence of carbon black had the opposite effect, reducing the abrasion of the R-Black sample.

Rebound resilience defines the polymer’s response following a rapid deformation due to impact. Because of its molecular structure, an elastomer can deform and return to its original shape. Part of the energy from the impact is returned as mechanical energy, while part is dissipated as heat. A greater rebound resilience indicates a higher elasticity. All the tested samples, the revulcanized and the reference V-NR samples, exhibited high rebound. Obtained rebound resilience was slightly lower for the revulcanizates. The average value for the revulcanized samples was 75.5%, ranging from 74.5% to 76.9%, while for V-NR it was 77.2%. The obtained values were typical for vulcanizates of NR-based blends without fillers. The rebound resilience value significantly decreases with a larger amount of filler. In studies by Li et al. [39], the resilience value of unfilled NR-based blends ranged from 79.0 to 80.5%, dropping to 40.5–43.4% after being filled with carbon black and silica. No similar effect was observed in the tested samples of the revulcanizates due to the low content of non-rubber additives. The lack of significant differences between the revulcanized samples and V-NR indicates that the revulcanized material in the tested range it did not show any significant signs of degradation.

3.5. Chemical Structure

The FTIR analysis results are presented in Figure 6. For comparative purposes, the analysis was conducted on samples without color additives, B-Clear, D-Clear, R-Clear, and V-NR. The NR sample was a flat film made from dried, non-crosslinked latex (LATZ).

Table 7. Most significant bands of the clear samples.

Wavenumber (cm−1)	Assignments
3200	-OH moisture	[40,41]
3040	C=C (stretching)	[40]
2850	C–H stretching of the CH3 group	[24,40]
2915	CH2 asymmetric stretching	[24,40]
2960	C-H stretching of aliphatic groups	[24,40]
1750	C=O	[40]
1660	C=C (stretching)	[40,42,43,44]
1590–1660	Nitrogen—connected with proteins	[40,41]
1540	C=O	[45]
1460	-CH2- scissor vibration typical signals of methyl and methylene groups (bending)	[24,40]
1380	-CH3 stretching vibration	[24,40]
1260	C-O	[44]
840	C=CH (stretching)	[44]

shows the most significant absorbance bands assigned to the bonds in the samples.

The primary peaks typical for methyl and methylene groups in the main chain are clearly visible in all spectra. In the range between 2770 and 3000 cm⁻¹, a characteristic group of peaks related to the stretching bonds in the saturated hydrocarbon backbone (2850, 2915, 2960 cm⁻¹) is evident. The peak at 1460 cm⁻¹ corresponds to the -CH₂ bond, while the peak at 1380 cm⁻¹ corresponds to the -CH₃ bond. The unsaturated bonds in the main chain can be observed as peaks at 1660, 840, and a small peak at 3040 cm⁻¹. The peak at 1660 cm⁻¹ is highest in the raw latex sample NR and gradually decreases in the subsequent samples: B-Clear (balloon), D-Clear (devulcanizate), and R-Clear (revulcanizate). Additionally, when comparing the V-NR and R-Clear samples, no differences were observed in the peaks associated with double bonds. In these samples, no peak characteristic of carbonyl groups was detected around 1750 cm⁻¹, which would indicate an adverse effect of devulcanization, such as oxidation of the polymer chain.

The broad peak around 3200 cm⁻¹ corresponds to the presence of -OH groups and gradually diminishes. This result may relate to trace moisture residue [46], as non-vulcanized latex films retain moisture more readily. Meanwhile, the waste sample (vulcanized latex) showed this peak less distinctly, and it disappeared entirely in the spectrum of the revulcanized sample due to the higher temperature.

The presence of peaks in the 1590–1660 cm⁻¹ range has been reported in studies on biodegradation and is associated with biofilm as a nitrogen source [40,41]. In the case of the NR and balloon waste samples, these peaks may indicate the presence of proteins found in natural latex.

The peak around 1260 cm⁻¹ appears with similar intensity only in the R-Clear and V-NR samples and is absent in the other samples. This band is often attributed to the -C-O bond [44] and may originate from zinc stearate. The 1540 cm⁻¹ peak in the R-Clear and V-NR samples may also be related to zinc stearate [45] formed after revulcanization and vulcanization of these samples. In the case of samples NR and B-Clear, the peak could be connected with ZnO and carboxyl group from stabilizers present in latex film (made of LATZ) and balloons.

3.6. Microstructure

SEM (Scanning Electron Microscopy) plays a key role in the analysis of rubber materials. Thanks to its high resolution and ability to image the surface of materials, SEM enables detailed analysis of the surface structure of rubber, which can help to understand its mechanical and physical properties. It allows detailed observations of the surface structure of rubber, including assessment of the degree of porosity, microcracks, irregularities, and the distribution and size of particles in the material. In the case of recycled materials, SEM can be used to analyze the effectiveness of the revulcanisation process by assessing the homogeneity of the newly formed polymer network and identifying areas where the process has not entirely run its course.

SEM images of studied materials are presented in Figure 7. The images were taken at 100x magnification for devulcanized and revulcanized samples, the reference sample comprising NR, and the vulcanized NR sample.

The images show differences in the surface structure of the samples before and after vulcanization. The reclaimed and unvulcanized NR samples from latex were more plastic, which can be visible as surface unevenness. All samples exhibited a relatively good level of uniformity and homogeneity, but some fragments of material with sharper edges can be seen in the images of the devulcanized samples. It may be the result of incomplete devulcanization. These are most apparent in the D-Red sample, which exhibited the lowest sol fraction (Table 4). Meanwhile, the surfaces of the D-White sample appear the most homogenous; this sample had the lowest Mooney viscosity (Table 2), indicating a better degree of devulcanization. This is also confirmed by the highest sol fraction of D-White (Table 4).

Comparing the surfaces of the vulcanized samples, similar patterns can be observed. On the surface of the R-Red sample, rough fragments with sharp edges are visible, which may indicate the worse homogeneity of this revulcanizate, reflected in the lowest sol fraction of the R-Red reclaim. The surface of the other revulcanizates is flatter and looks similar, which may indicate better quality of the obtained material.

Smaller particles are noticeable on all samples, which may be talc particles that were present on the surface of all samples (even the NR sample was produced using a dipping method with a talc-containing coagulant).

4. Conclusions

The study focused on waste material from pre-vulcanized NR latex (PV latex) balloons. The waste material was sorted by color. Mechanical devulcanization of the waste material was performed, followed by revulcanization of the mixtures obtained from the devulcanizates and sulfur-based curing system. The reference sample was prepared from NR obtained from dried concentrated latex.

The revulcanizate exhibits better mechanical and tensile properties (higher hardness and tensile strength) and lower swelling compared to the reference sample V-NR. The higher torque increment (ΔM) during the RPA test has resulted in the higher crosslink density of the revulcanizate samples than the reference vulcanizate of virgin rubber V-NR. The results may have been influenced by the content of coloring additives present in the balloons (titanium dioxide, carbon black) and the residue from the original latex prevulcanising system. SEM analysis revealed some differences in the morphological structure of the examined samples; however, these did not significantly affect the deterioration of their tensile properties.

On the other hand, the revulcanized samples showed some signs of degradation, achieving slightly lower abrasion resistance.

In summary, based on the obtained results, it can be concluded that the tested waste material has a high potential for material recycling, which encourages further testing. A significant advantage of the studied waste is its relatively high potential for future revulcanization. This is primarily due to the low crosslink density of the balloon rubber. The latex coloring agents were shown to have only a minor impact on the results obtained for the individual revulcanizates. The high quality of the polymer is also related to the product’s specifications. Firstly, the rubber is derived from high-quality natural latex. Secondly, the manufacturer must ensure that the balloons meet the chemical safety standards for toys from regulations and norms, like the Toy Directive or EN 71.

The results suggest that devulcanizate could be suitable for applications such as composites. Blends containing natural rubber (NR), which imparts the necessary elasticity to the components, are used in various fields, including construction [47]. The low crosslink density of waste from latex balloon production supports the substitution of raw rubber with devulcanizate in such blends. Research in this area would align with efforts aimed at the sustainable management of natural resources and a closed-loop waste economy.