1. Introduction
In recent years, vegetable oils and waste cooking oils have gained attention as cost-effective, renewable resources for the production of fatty alcohols. These alcohols are key components in the manufacturing of plasticizers, cosmetics, surfactants, lubricants, and diesel-range alkanes, which are commonly referred to as second-generation biodiesel. Dodecanoic acid, comprising approximately 50% of palm kernel oil and coconut oil, is a typical bio-based aliphatic fatty acid with a medium carbon chain length (C12). Globally, approximately 3900 thousand tons of fatty alcohols were consumed in 2022, with demand expected to grow steadily at a rate of 4.74% from 2022 to 2030 [
1]. Currently, commercial catalysts such as metal sulfides (e.g., NiMoS
4 and CoMoS
4) and copper chromite (CuO/CuCr
2O
4) are employed for the hydrogenation of natural oils to produce diesel-range alkanes or fatty alcohols [
2]. However, the use of these catalysts is constrained by the harsh reaction conditions (200–400 °C and high H
2 pressures), as well as the toxic nature of sulfur and chromium compounds, which limit their broader application. This has spurred ongoing research to develop more efficient heterogeneous catalyst systems for fatty acid hydrogenation. Homogeneous catalysts such as sulfuric acid, potassium hydroxide, hydrochloric acid, sodium hydroxide, and sodium methoxide are frequently used due to their high activity and low cost. Nevertheless, these catalysts are non-recoverable, which poses challenges for industrial applications. Additionally, they cause corrosion in equipment and lead to environmental concerns during the purification of end products, which necessitates multiple washing stages [
3]. As a result, the search for alternative heterogeneous catalysts, such as metal oxides, has been intensifying. Previous research has shown that noble metal (Pt, Ru, Rh, Pd, Au, and Ir)-supported various metal oxides can act as effective catalysts for fatty acid hydrogenation. Despite this, achieving high selectivity for desired products remains a challenge, necessitating further refinement of catalyst design.
The hydrogenation reaction typically saturates C=C, C=O, and C–O–C bonds in the presence of metal catalysts such as Cu, Ni, Rh, Ru, Pd, and Pt. In the pursuit of improved efficiency, heterogeneous catalysts are being developed to replace conventional homogeneous systems. Commercial fatty alcohol production predominantly utilizes Cu–Cr-based catalysts under stringent conditions (250–350 °C, 10–30 MPa of H
2). This is due to the inherent difficulty of reducing fatty acids [
4,
5,
6]. However, the toxicity and environmental impact of chromium-based catalysts present significant obstacles to their widespread use. Catalysts composed of noble metals such as Ru and Pt have shown promise in the selective hydrogenation of fatty acids. Among various Ru-based catalysts [
7], the bimetallic Ru–Sn/Al
2O
3 system has demonstrated the highest selectivity for fatty alcohols, operating at 250 °C and 5.6 MPa of H
2 pressure [
8]. Moreover, replacing Al
2O
3 with TiO
2 as the support for Ru–Sn catalysts has enhanced both activity and selectivity due to favorable interactions between the Ru-Ti metal–support pair. Some studies have explored the use of bimetallic catalysts in the hydrogenation of fatty acids. For instance, Ru–Sn catalysts supported [
9] on various substrates such as Al
2O
3, SiO
2, ZrO
2, and TiO
2 have been employed to hydrogenate fatty acids at 300 °C and 6 MPa of H
2 pressure, achieving over 95% conversion and approximately 80% selectivity to dodecanol. However, the high reaction temperature remains a limitation. In contrast, hydrogenation of decanoic acid using Pt–Re catalysts supported on TiO
2 at 130 °C and 2 MPa for 4 h resulted in 79% conversion and 75% selectivity to decanol. Similarly, at 300 °C and 3 MPa, octanoic acid was hydrogenated with Ru–Sn/ZnO catalysts to produce octanol with over 99% conversion and 93% selectivity [
10].
Further investigations have been conducted previously, and they found that adjusting catalyst supports and/or adding a second metal can improve the selectivity and conversion in the hydrogenation reactions. For instance, crotonaldehyde was converted to crotyl alcohol using an Ir/SiO
2 catalyst at 30 °C and 0.8 MPa hydrogen pressure over 1 h, although the conversion rate was relatively low at 10.9% [
11]. The authors have achieved modification of Ir/SiO
2 combined with metal oxides such as MoO
x and ReO
x, affording approximately 95% selectivity in the hydrogenation of crotonaldehyde at 30 °C and 8 MPa of H
2 for 1 h [
11]. In another study, the addition of Re to an Ir/SiO
2 catalyst facilitated the production of 1,3-propanediol from glycerol under reaction conditions of 120 °C and 8 MPa hydrogen pressure for 24 h [
12]. Despite these modifications, the conversion rate reached only 69%, with product selectivity remaining limited at 47%. Moreover, Ir–Re catalysts supported on activated carbon have been shown to hydrogenate adipic acid with 77% conversion and 52% selectivity under similar reaction conditions [
13], although these catalysts still require high H
2 pressure. During the hydrogenation of succinic acid over the Cu–Pd active center, the selection of metal oxide as a support material was significantly contributed to its product selectivity to γ-butyrolactone (by TiO
2), 1,4-butanediol (by hydroxyapatite and SiO
2), or tetrahydrofuran (by γ-Al
2O
3) at the same reaction conditions: 200 °C, 8 MPa H
2, and 96 h [
14,
15]. These findings emphasize the importance of the combination of the active metal element and catalyst support in the development of efficient catalyst systems for achieving high activity and selectivity in hydrogenation processes.
In this study, first of all, the authors revisit the catalytic performance in the hydrogenation of dodecanoic acid over various supported metal (Au, Pt, Cu, Ir, Ru, Pd, Mo, In, and Ni) catalysts using AlOOH as a support material, particularly for the selective production of dodecanol. Earlier reports described that metal-supported AlOOH catalysts possessed attractive performance not only for the hydrogenation [
16,
17] but also for various organic reactions [
18,
19]. After evaluating the catalytic activity, iridium (Ir) was selected as the optimal candidate for further screening. Secondly, the results of the support survey indicated that the Ir-based catalyst had a unique feature in the selective hydrogenation of fatty acids. It was particularly interesting that, when dodecanol was identified as the primary product across most catalysts, the Ir/Nb
2O
5 catalyst showed the highest yield (82.2%) at 200 °C and 8 MPa of H
2 for 17 h, whereas the Ir/MoO
3 catalyst produced dodecane as the main product with a significant yield of 90.7%. Finally, the findings indicated that the simple co-impregnation of Ir onto the Nb
2O
5 and MoO
3 could selectively facilitate the hydrogenation of fatty acids to the corresponding alcohol for chemical an alkane for fuel, respectively. To the best of our knowledge, this is the first report of a heterogeneous Ir-based catalyst for applying the selective hydrogenation of fatty acids. Although the Ir element is expensive and one of the rare metals, these findings hope to open new possibilities of heterogeneous Ir nanoparticle-mediated hydrogenation reactions, particularly for the utility of wasted fatty acids.
2. Results and Discussion
The performance of different metals supported on AlOOH for the hydrogenation of dodecanoic acid to dodecanol was investigated, as listed in
Table S1 of the Supplementary Materials (SM). Among the nine elements of Ir, Pt, Ru, Cu, Ni, Pd, In, Au, and Mo monometallic catalysts, the highest yield of dodecanol (52.2%) was obtained over the Ir-based catalyst with 72.5% conversion. The Pt-based catalyst gave a 27.1% yield with 56.0% conversion, and the Ru- and Cu-based catalysts showed slight yield values of 3–5%. In contrast, other elements such as Ni-, Pd-, In-, Au-, and Mo-based catalysts exhibited poor performance with low conversion (<25.0%) and no dodecanol formation (0% yield). As a result, the Ir was selected as the metal center for further investigations of metal-supported catalysts in the hydrogenation of dodecanoic acid. Previously, there were some reports for the hydrogenation reaction over heterogeneous monometallic Ir catalysts [
11,
20,
21,
22,
23,
24,
25], particularly Ir/SiO
2; therefore, focusing on the Ir-based catalyst in the hydrogenation of fatty acids is an attractive subject.
To obtain insight into the role of Ir metal, eight types of catalyst support were prepared and tested for the hydrogenation of dodecanoic acid. The results are shown in
Figure 1. The details of the activity are listed in
Table S2 of the SM. The catalytic activity of the various monometallic Ir-based catalysts for the hydrogenation of dodecanoic acid exhibited a wide range. There were mainly three categories, i.e., dodecane formation, highly-selective dodecanol production, and inactive, among the eight types of Ir-based catalyst performances. Notably, under the same reaction conditions outlined in
Figure 1, quantitative conversions of dodecanoic acid (100% conversion) were achieved with Ir-supported MoO
3, Ta
2O
5, and Nb
2O
5, while poor activities (<10% conversions) were observed with Ir-supported ZrO
2, SiO
2, and WO
3. Among the catalysts tested, the Ir/Nb
2O
5 catalyst demonstrated the highest yield of dodecanol (82.2%) and complete conversion of dodecanoic acid, highlighting its considerable potential for the sustainable production of fatty alcohols from bio-based sources. Ir-supported AlOOH and TiO
2 (anatase) also gave good yields for the dodecanol production as 52.2% and 48.7%, respectively. Under identical conditions, interestingly, the Ir/MoO
3 catalyst also achieved complete conversion (100%), with dodecane, constituting 90.7% of the main product, with low yield towards dodecanol production (9.3%). The Ir/Ta
2O
5 gave both capacity for the production of dodecanol with 39.5% yield and 60.5% yield with 100% conversion of dodecanoic acid. Accordingly, the types of support materials for the Ir-based catalysts could alter the reactivity, and Nb
2O
5 and MoO
3 gave a significant trend with high selectivity for dodecanol and dodecane, respectively, at the same reaction conditions. Additionally, it was very interesting that the yield to dodecane decreases drastically at lower reaction temperature (170 °C) and lower H
2 pressure (4.0 MPa) over the Ir/MoO
3; it decreased from 90.7% to 5.7%, while Ir/Nb
2O
5 maintained a good yield for dodecanol (89.1% yield) under identical conditions. It indicated that optimizing not only the catalyst support material but also reaction conditions was a crucial factor in the Ir-based catalyst for the selective hydrogenation of fatty acids.
Time-dependent reaction profiles of the hydrogenation of dodecanoic acid over the Ir/Nb
2O
5 (at 8.0 MPa H
2) and Ir/MoO
3 (at 4.0 MPa H
2) catalysts at 200 °C are shown in
Figure 2. Note that the H
2 pressure for Ir/MoO
3 was decreased to 4.0 MPa because its reactivity is so significant, which allowed us to easily trace the reaction profile. As shown in
Figure 2a, over the course of the reaction with the Ir/Nb
2O
5 catalyst, the conversion of dodecanoic acid increased steadily from 8 to 17 h. The conversion remained constant at 100% for up to 40 h of reaction time. The yield of dodecanol reached its maximum yield of 86.2% at 24 h; however, it progressively declined after 40 h to 58.9% yield. This decrease inversely correlated with a corresponding increase in the yield of dodecane from 13.8% to 41.1%. Therefore, the production of dodecane occurred via dodecanol stepwise. Under the milder H
2 pressure condition at 4.0 MPa, the conversion of dodecanoic acid over the Ir/MoO
3 catalyst gave the same profile: an increase and then decrease trend of dodecanol production with a linear increase of dodecane yield, as shown in
Figure 2b. Complete conversion was achieved within 40 h, with a high yield of dodecane (85.3%), a saturated aliphatic alkane, emerging as the dominant product. Note that the performance of Ir/MoO
3 was very significant and gave 91.3% yield for dodecane at 6 h under 8.0 MPa of H
2. The results are summarized in
Table S3 of the SM. Accordingly, the Ir/MoO
3 catalyst led to the activation of more reactant molecules much faster than the Ir/Nb
2O
5 catalyst to dodecane. To emphasize the good yield for dodecanol production over Ir/Nb
2O
5, time-dependent reaction profiles of the hydrogenation of dodecanoic acid over Ir/Nb
2O
5 under the milder reaction conditions (at 170 °C and 4.0 MPa H
2) were further examined; however, the yield of dodecanol peaked at 17 h at 89.1% yield, after which it progressively declined with extended reaction time (see
Table S3 of SM).
The 1.0 g scale of reaction (10 times larger scale) was also compared over Ir/Nb
2O
5 and Ir/MoO
3 catalysts (see
Table S4 of SM). The Ir/Nb
2O
5 catalyst predominantly produced dodecanol as the major product, achieving a high conversion rate of 99.8% with 68.2% yield of dodecanol and 30.5% yield of dodecane even at 40 h of reaction. In contrast, the Ir/MoO
3 catalyst demonstrated an exceptional yield of dodecane, approximately 91.7%, with complete conversion (100%) at 24 h of reaction. At this time, GC-TOFMS analysis revealed the presence of dodecyl laurate, which can be synthesized from the substrate (dodecanoic acid) and the alcoholic product (dodecanol) through esterification, as illustrated in
Scheme 1. Therefore, figuring out how to prohibit the side-reaction of the esterification is crucial for the application of a large-scale reaction to afford the corresponding alcoholic product with high yield in the hydrogenation of fatty acids. The utility of the ester form as the starting material is a well-known approach [
9,
26]; however, large-scale applications still seem to be challenging.
In the current study, Nb
2O
5 was selected as the catalyst support due to its reported superior performance in facilitating the selective hydrogenation of dodecanoic acid to the alcoholic form (dodecanol). To evaluate the stability and reusability of the catalyst, the Ir/Nb
2O
5 system was subjected to a straightforward regeneration process. The catalyst was centrifuged, thoroughly washed with ethanol solvent, and subsequently dried under vacuum before being employed in subsequent reaction cycles.
Figure 3 demonstrates that Ir/Nb
2O
5 catalyst could maintain its high activity and good selectivity for dodecanol in the hydrogenation of dodecanoic acid even after the fifth reuse. Detailed values are listed in
Table S5 of the SM. Note that the recycling test was examined at H
2 (4.0 MPa), 170 °C, and 17 h, which was the best condition for Ir/Nb
2O
5 to afford high yield for dodecanol (see
Table S3 of the SM). This stability implied the strong interaction between metallic iridium and the Nb
2O
5 support, which is crucial for the efficient conversion of dodecanoic acid to dodecanol. However, the selectivity toward dodecanol exhibited some fluctuations across the recycling runs.
To evaluate the versatility of the catalytic activity, the scope of hydrogenation using the Ir/Nb
2O
5 catalyst was extended to a range of other aliphatic carboxylic acids, including hexadecenoic acid (C
16H
30O
2), tetradecanoic acid (C
14H
28O
2), and oleic acid (C
18H
34O
2), with the results summarized in
Table S6 of the SM. In all cases, the catalyst exhibited high activity, achieving complete conversion (100%) across all three substrates. The catalytic hydrogenation of aliphatic carboxylic acids consistently demonstrated the highest selectivity toward the corresponding alcohols. The production yields of hexadecanol and oleyl alcohol were, respectively, obtained as 80.8% and 78.0% at 200 °C in 8.0 MPa H
2. Although the production of tetradecanol was 52.0% yield with the corresponding alkane (48.0%), the optimized reaction conditions, including temperature, pressure, and hydrogen availability, would arrange the partial reduction of the carboxyl group to alcohols rather than complete hydrogenation to alkanes.
Powder XRD patterns for the Ir/Nb
2O
5 and Ir/MoO
3 catalysts are presented in
Figure 4, comparing the calcined and reduced catalysts. No significant differences were observed in the peak positions of metallic iridium among the catalysts; however, variations in intensity were significant. Both Ir/Nb
2O
5 and Ir/MoO
3 catalysts exhibited similar patterns across all calcined samples with their respective calcined supports themselves. Following reduction, metallic Ir peaks became prominent in the Ir/Nb
2O
5 catalyst, contributing to its enhanced performance in the hydrogenation of dodecanoic acid. In contrast, the Ir species in the Ir/MoO
3 catalyst were not clearly discernible after reduction, likely due to their smaller size and higher dispersion on the MoO
3 support, with only a faint peak observed at 2
θ = 41.7°. Additionally, MoO
2 species were distinctly visible in the XRD pattern of Ir/MoO
3 at 2
θ = 25.9° and 37.2°, corresponding to the (110) and (200) planes, respectively, with intensities increasing post-reduction.
To further validate the differences between Ir/Nb
2O
5 and Ir/MoO
3, TEM analysis was conducted to investigate the particle distribution and morphology of the catalysts. Iridium species have been effectively dispersed on the surfaces of Nb
2O
5 and MoO
3 supports, as illustrated in
Figure 5, where numerous bright black metal nanoparticles are observed on both Ir/Nb
2O
5 and Ir/MoO
3 catalysts. The Ir/Nb
2O
5 catalyst exhibits a particle size distribution ranging from 1.4 to 3.0 nm and mean diameter is to be 2.38 nm (
Figure S1 of SM). In contrast, the Ir/MoO
3 catalyst displays a large number of fine particles, with an average size that cannot be accurately determined because the distinctions between Ir and MoO
x support were difficult to identify. In the H
2-TPR profile of calcined samples, both Ir/Nb
2O
5 and Ir/MoO
3 gave a reducing peak at around 200–300 °C, and the H
2 consumption amount of Ir/MoO
3 (6.43 mmol g
−1) was much larger than that of Ir/Nb
2O
5 (0.43 mmol g
−1), as shown in
Figure S2 of the SM. This also suggested the presence of a unique interaction between Ir and Mo, which would be generated after the reduction process under an H
2 flow at 500 °C.
Figure 6a–d present the XPS analysis of the Ir 4f region results for the reduced and used catalyst samples in Ir/Nb
2O
5 and Ir/MoO
3, respectively. The deconvolution of the Ir 4f peaks revealed the presence of two distinct iridium species on the catalyst surface.
Figure 6a,b for the Ir/Nb
2O
5 illustrate the two doublets corresponding to Ir 4f
5/2 and Ir 4f
7/2. The binding energy (BE) for the metallic iridium species in the Ir 4f
7/2 orbital of Ir/Nb
2O
5 is observed at 60.8 eV, while a value of 63.8 eV is recorded for the Ir 4f
5/2 orbital. Differences are observed as +0.3 eV between the (a) fresh and (b) used Ir/Nb
2O
5 catalysts in the Ir 4s region. In contrast, the Ir/MoO
3 catalyst displays iridium oxide species (Ir
4+) as the shoulder peaks in both (c) reduced and (d) spent catalyst. Additionally, there is evidence of coordination between Ir–O and C=O species in the O 1
s region, as well as interactions between Ir and oxygen-rich C 1
s. However, differences before and after the reaction showed no trends in the Ir states. One can consider that partial oxidation by air during sample transportation is the reason for its difficulty in analysis. Importantly, there were significant differences between Ir/Nb
2O
5 and Ir/MoO
3; the former was mainly composed of the Ir metallic state, whereas the latter contained a mixture of Ir metal and oxides. The deconvolution of the Mo 3d peaks for the Ir/MoO
3 catalyst after reduction revealed the presence of Mo
6+ species (232.5 eV and 235.7 eV) and Mo
4+ species (229.7 eV and 233.0 eV), as shown in
Figure S3 of the SM. Both XRD and XPS analyses indicated a transition from the MoO
3 (Mo
6+) to MoO
2 (Mo
4+) phase during this process. Accordingly, the presence of reduced Mo species facilitates the formation of oxygen vacancies and/or Ir–Mo alloying, thereby contributing to an enhancement in catalytic activity. The importance of these factors on the hydrogenation reaction has been discussed in previous papers [
27,
28]. Interestingly, Ir and Mo combinations are likely attracted to many researchers for hydrogenolysis and hydrodeoxygenation reactions over heterogeneous catalysts by pressurized H
2 [
29,
30].