Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2024 Sep 9;14:20925. doi: 10.1038/s41598-024-71790-9

Ortho to para hydrogen conversion over bimetallic iron and cobalt catalysts

Liujing Yang 1,2, Xinbao Li 1,✉, Ying Chen 2, Xiaoling Zheng 2, Kai Sun 2
PMCID: PMC11383857  PMID: 39251648

Abstract

The catalytic conversion of ortho-hydrogen (o-H2) to para-hydrogen (p-H2) serves as a crucial step in the storage of liquid hydrogen. A variety of iron-cobalt bimetallic catalysts (FCO) were synthesized using a precipitation method, incorporating diverse levels of Co doping into Fe-based catalysts. The effects of Co doping on the crystal structure, porosity, and magnetism of FCO catalysts were studied by XRD, N2 physical adsorption, FTIR, XPS, and VSM analyses. The efficiency of ortho-para hydrogen conversion over FCO at 77 K was evaluated. It was found that the catalyst’s lag coefficient was significantly improved by Co doping, leading to an increase of magnetic moment. The catalyst of FCO-5 with a Fe/(Fe + Co) molar ratio of 0.5 exhibited the highest activity in ortho-para hydrogen conversion. The corresponding conversion rate, outlet p-H2 content, and the reaction rate constant were 99.2%, 49.7% and 291.7 mol·L−1·s−1, respectively, under a gas hourly space velocity of 5400 h−1.

Subject terms: Chemistry, Energy science and technology

Introduction

Hydrogen, with diverse sources and environmentally-friendly combustion characteristics1, is considered the most promising clean energy within the global decarbonization framework2,3. Considering the potential of hydrogen energy, numerous countries and organizations have introduced strategic plans for its advancement. Notably, the European Union’s “EU Hydrogen Strategy”4 and the U.S. Department of Energy’s (DOE) “Hydrogen Shot” initiative introduced in 20215. The storage and transportation of hydrogen, pivotal elements in the hydrogen energy sector, present challenges to its advancement. Liquid hydrogen storage technology, with its high energy density, is particularly well-suited for long-distance transportation6–8. Furthermore, low-temperature liquid hydrogen storage provides enhanced safety measures compared to high-pressure storage9. Nevertheless, the conversion of ortho-hydrogen (o-H2) to para-hydrogen (p-H2) during hydrogen liquefaction results in the release of heat due to their differing physical properties10,11. Thus, for industrial purposes, the para-hydrogen (p-H2) content in liquid hydrogen storage should exceed 95%12.

Hydrogen gas is categorized into ortho-hydrogen and para-hydrogen based on nuclear spin orientation13. Under equilibrium conditions, normal hydrogen (n-H2) at room temperature and above consists of 75% o- H2 and 25% p-H214. With decreasing temperature, o-H2 undergoes a gradual conversion to the lower spin state of p-H2, with the process occurring slowly15. The production and storage of liquid hydrogen require low-temperature conditions11,16. The gradual conversion of o-H2 to p-H2 leads to losses in liquid hydrogen storage17. Consequently, the utilization of catalysts is crucial for the efficient conversion of o-H2 to p-H2, ensuring effective long-term liquid hydrogen storage.

With a rapid development of catalytic ortho-para hydrogen conversion (O-P conversion) in industry, a lot of reaction mechanisms were proposed for O-P conversion. The theory proposed by Wigner in 1993 was well recognized, in which the important of paramagnetic catalysis in molecular hydrogen O-P conversion was reported18. The catalyst’s paramagnetic center significantly accelerated the O-P conversion. The O-P conversion rate depended on paramagnetic catalysts and proton magnetic moments. Furthermore, in 1973, Petzinger and Scalapino reported that the O-P conversion rate was dramatically boosted by augmenting the catalyst’s magnetic moment and decreasing the distance between reactants and H2 surfaces19. Thus, these represent the two fundamental approaches in developing O-P conversion catalysts. Transition metals which contain unpaired electrons, coupling with strong magnetic moments, would be preferred for O-P conversion20. Iron was extensively employed in industrial synthesis of O-P conversion catalysts due to its abundant availability and low cost21.

Svadlenak et al.22 conducted a study on various iron-zinc compounds, encompassing γ-Fe2O3, for O-P conversion at 78 K. The findings revealed that γ-Fe2O3 demonstrated the highest catalytic activity, with α- Fe2O3, α- Fe2O3-ZnO, and ZnFe2O4 showing progressively lower catalytic activity. The catalysts exhibited a range of magnetic properties including ferromagnetic, antiferromagnetic, weak antiferromagnetic, and paramagnetic with subtle ferromagnetism. The study conclusively demonstrated that ferromagnetic arrays are superior to antiferromagnetic arrays in promoting hydrogen nuclear spin conversion. Das et al.23 synthesized LaFeO3/Al2O3 catalysts via the citrate sol–gel technique and observed that the catalyst attained the highest spin conversion rate post-activation through calcination at 773 K. At 17 K, a catalyst with La:Fe ratio of 2:8 (20La0.2Fe0.8O3/Al2O3) achieved a remarkable 99.8% conversion of o-H2 to p-H2 within a 120 min duration. Karlsson24 examined the commercially available porous particle catalyst, IONEX®, predominantly comprising iron oxide (Fe2O3). The conversion process involves spin conversion on magnetically aligned surfaces, with the catalyst’s magnetism emanating from iron and possessing a magnetic moment of 5.92 Bohr magnetons. The results showed that achieving a 99.8% conversion of normal hydrogen into para-hydrogen at 16 K took merely 80 min. In addition, Xu et al.25 employed ammonia hydroxide (NH4OH) as a precipitant to prepare FeCo catalysts through a precipitation method. They found that the catalyst with a Co/Fe ratio of 3:7 achieved the highest outlet 38.5% p-H2 content under a H2 flow rate of 500 ml·min−1.

Polyukhov et al.26 showcased the viability of Metal–Organic Frameworks (MOFs) as catalysts for O-P hydrogen conversion, introducing efficient catalysts such as M-MOF-74 (M = Mn, Co, Cu, Ni, Zn). Among these catalysts, Ni-MOF-74 displayed an O-P conversion rate constant (k) of 26,000 min−1 g−1 at 77 K, suitable for use at temperatures lower than 15 K. Despite the excellent performance of MOF catalysts, their complex and costly preparation process necessitates the proposal of a straightforward, cost-effective production method for these catalysts. Compared to the hydrothermal method27 and integrated catalytic method28, the precipitation method has the advantage of a simple preparation process.

This study synthesized a series of FCO catalysts doped with Co elements using a precipitation method. The performance of the FCO catalyst in converting ortho-para hydrogen at 77 K was evaluated by adjusting the Fe/(Fe + Co) molar ratio to identify the optimum FCO ratio. The influence of Co doping on Fe-based catalysts was analyzed through XRD, XPS, and VSM techniques. The impact of catalyst micro-porosity and surface area on catalytic performance was investigated using N2 adsorption–desorption tests. The successful combination of Fe and Co was confirmed using FTIR technology.

Experimental methods

Material synthesis

All reagents used are of analytical grade. Ferric nitrate nonahydrate (Fe(NO3)3·9H2O) and Cobalt(II) nitrate hexahydrate (Co(NO3)2·6H2O) are the precursors for iron and cobalt respectively. NaOH is used as the precipitating agent.

To synthesize a series of Fe–Co catalysts using the precipitation method, FCO-9 is taken as an example. Dissolve 18 g of Fe(NO3)3·9H2O in 100 ml of H2O, then sonicate for 10 min (solution A). Dissolve 1.5 g of Co(NO3)2·6H2O in 100 ml of H2O, then sonicate for 10 min (solution B). Slowly pour solution A into solution B to obtain solution C. Dissolve 5.6 g of NaOH in 100 ml of H2O (solution D), then slowly add it dropwise into solution C with stirring until the pH reaches 8. Continue stirring the mixture for 30 min, then allow it to age at room temperature for at least 6 h. Filter and wash the precipitate with deionized water until the pH of the filtrate reaches 7. Dry the filter cake overnight in a drying oven, then crush the resulting catalyst to a particle size of 40–60 mesh. The molar ratios of n(Fe)/n(Fe + Co) were 0.9, 0.7, 0.5, 0.3 and 0.1, and they were named as FCO-9, FCO-7, FCO-5, FCO-3, and FCO-1, respectively. Pure phase samples were prepared according to the above process without Co(NO3)2·6H2O and without Fe(NO3)3·9H2O, and named FO and Cob, respectively.

Materials characterization

Structural analysis of Fe–Co samples was conducted using X-ray diffraction (XRD, Smartlab 9 kW) with Cu Kα radiation. Nitrogen adsorption–desorption experiments at 77 K were performed using a Micromeritics ASAP 2460 analyzer. The surface area, total pore volume, and average pore size of the samples were calculated using the Brunauer–Emmett–Teller (BET) method, while the Barrett-Joyner-Halenda (BJH) method was employed for pore size distribution analysis. The microstructure and morphology of the sample were studied using a scanning electron microscope (SEM, Hitachi S4800). Performing compositional homogeneity analysis at the nanoscale using energy dispersive spectroscopy (EDS). Fourier transform infrared spectroscopy (FTIR, Thermo Scientific Nicolet iS20) was used for qualitative and quantitative analysis of functional groups and chemical bonds in the samples. X-ray photoelectron spectroscopy (XPS) analyses were performed using the Thermo K-alpha spectrometer, while magnetic measurements were carried out at a temperature of 77 K using the PPMS-9 T dc magnetometer from American Quantum Design.

Catalytic performance at ultra-low temperatures

The sample was tested using a fully automated O-P conversion and reduction device. Firstly, the sample (mass = 1 g) was placed in a furnace for activation. The activation conditions were vacuum, with a temperature of 130 °C for 6 h. After activation, the reaction vessel containing the sample was placed in a liquid nitrogen (77 K) low-temperature bath. Normal hydrogen was deoxygenated and dehydrated before being introduced into the reaction vessel for the reaction. The normal hydrogen flow rate was adjusted using a mass flow meter, and the content of para-hydrogen after catalysis was measured using a gas chromatograph (GC-9790II, Fuli Instruments) equipped with a 5A molecular sieve packed column and a thermal conductivity detector (TCD).

In this work, measurements are taken every 4 min for 6 consecutive times when the hydrogen flow rate is stable and the data is smoothed, and the average value is calculated.

Results and discussions

XRD

X-ray diffraction (XRD) was utilized for the compositional and crystal structure characterization of the samples, with results presented in Fig. 1. The analysis of Fig. 1 reveals that Co doping altered the crystalline phases of the samples and enhanced their crystallinity. All diffraction peaks of Cob were found to align with those of Co(OH)2 (JCPDS No.89-8616), demonstrating good crystallinity. The prominent diffraction peaks observed at 2θ = 35.63° and 62.93° in the FO samples were attributed to Fe2O3 (JCPDS No.39-1346), identifying it as the primary crystalline phase29. Introduction of Co led to the emergence of new phases in the FCO series of samples. Unique diffraction peaks corresponding to CoFe2O4 (JCPDS No.03-0864) were detected in FCO-9, FCO-7 and FCO-5 (marked by asterisks). Furthermore, the characteristic diffraction peaks of Fe2O3 vanished in the FCO-7 sample. As the Co content increased, characteristic diffraction peaks of Co3O4 (JCPDS No.43-1003), Fe3O4 (JCPDS No.26-1136), and CoO(OH) (JCPDS No.07-0169) were observed in FCO-5, FCO-3, and FCO-1 samples, respectively. The XRD results indicated a trend towards transformation into Co(OH)2 with increasing Co doping. In FCO-1, the CoO(OH) phase displaced the coexistence of Co3O4 and Fe3O4, resulting in the formation of new crystal facets. Additionally, a novel peak (311) emerged at 2θ = 35.45° in FCO-9. Incrementing Co content led to the appearance of new crystal facets (003), (220), (222), (400), (015), (511), and (113) at 2θ = 20.24°, 31.27°, 38.55°, 44.81°, 50.58°, 59.35°, and 69.17° in FCO-5. Conversely, in FCO-1, new crystal facets (012), (104), and (110) appeared at 2θ = 38.89°, 45.86°, and 65.34°, respectively. Notably, as Co content increased, the intensity of diffraction peaks rose, and the peak shape sharpened, indicating an enhancement in crystallinity due to Co doping25,30.

Fig. 1.

Fig. 1

Open in a new tab

XRD pattern of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

The grain sizes of Co species in the samples were calculated using the Scherrer formula, and the results are shown in Table 1. It was observed that FCO-9 and FCO-7 did not exhibit distinct characteristic diffraction peaks of Co species. Meanwhile, the grain sizes of CoOOH and Co3O4 decreased sequentially in FCO-5, FCO-3, and FCO-1.

Table 1.

Initial composition and grain size of CoOOH and Co3O4 for FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

Catalyst Fe/(Fe + Co) (%) CoOOH (nm) Co3O4 (nm)
FO 100 / /
FCO-9 90 N.A N.A
FCO-7 70 N.A N.A
FCO-5 50 20.1 23.9
FCO-3 30 21.3 16.8
FCO-1 10 14.8 13.0
Cob 0 / /
Open in a new tab

N.A. stands for “not available,” indicating that data cannot be provided due to very small grain size.

N2 physical adsorption

In order to investigate the effect of Co doping and different Fe/Co ratios on the surface area and pore structure of the samples, N2 adsorption and desorption tests were conducted. Figure 2a shows the N2 adsorption–desorption isotherms of samples. All samples exhibit type IV isotherms, indicating a typical mesoporous structure. With the increase of Co content, the hysteresis loop transitions from an H2 type to an H4 type, indicating a transformation from “ink-bottle” pores to slit-like pores. This suggests that the variation in Co doping levels brings about different pore structures in the samples. Combining with Fig. 2b, the pore size distribution of FCO-9 is mainly concentrated between 1.7 and 4.9 nm, with smaller pore sizes leading to the formation of “ink-bottle” pores at relative pressures between 0.4 and 0.631. Therefore, the isotherm exhibits a typical H2 type hysteresis loop. The saturated adsorption plateau at the high-pressure stage indicates a relatively uniform pore size distribution for FCO-9. As the Co content increases, FCO-5 starts to exhibit an H3 type hysteresis loop, which is typically associated with slit-like pores formed by the stacking of sheet-like particles. The results calculated according to the lag coefficient (equation S5) are shown in Table 2. The results indicate that as the Co content increases, the N2 adsorption decreases continuously. The lag coefficient first increases and then decreases, reaching the highest value at FCO-5. This indicates that FCO-5 has a smaller degree of pore openness, allowing for better gas interaction. In the P/P0 > 0.8 range, the adsorption rate for FCO-1 continuously increases without significant adsorption limitation, due to capillary condensation occurring in the pores. This implies the presence of larger and diverse pore types within FCO-1. The pore distribution (Fig. 2B) was determined using the Barrett-Joyner-Halenda (BJH) method from the N2 adsorption branch of the isotherms. The pores of FO and Cob are mainly composed of microspores (< 2 nm) and macrospores (> 50 nm), while the FCO series are mainly composed of mesoporous (2–50 nm). Among them, FCO-9, FCO-7, and FCO-3 contain a small amount of microspores, while FCO-1 contains a small amount of macrospores. FCO-5, FCO-3, and FCO-1 feature a slit-type pore structure. Compared to “ink bottle” pores (FCO-9 and FCO-7), these slits are smaller, hindering gas flow through the pores and facilitating thorough catalyst reaction with hydrogen. Among samples primarily featuring slit-type pores (FCO-5, FCO-3, and FCO-1), FCO-5 has the largest BET surface area, providing enough contact between n-H2 and FCO-5. Additionally, FCO-5 exhibits the highest lag coefficient, indicating an extended contact time with H2. Therefore, considering both pore structure and BET surface area, FCO-5 shows the most effective ortho-para hydrogen conversion performance.

Fig. 2.

Fig. 2

Open in a new tab

(a) Nitrogen adsorption–desorption isotherms and (b) pore size distribution graphs of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

Table 2.

Surface area, pore volume, pore size and lag coefficient of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

Catalyst FO FCO-9 FCO-7 FCO-5 FCO-3 FCO-1 Cob
BET Surface Area (m2·g−1) 256.76 267.96 242.40 137.36 129.97 89.98 180.54
Pore Volume (cm3·g−1) 0.22 0.19 0.22 0.27 0.19 0.26 0.30
Pore Size (nm) 3.49 2.79 3.67 7.95 5.87 11.7 6.56
Lag Coefficient 0.005 0.006 0.025 0.075 0.070 0.064 0.015
Open in a new tab

FTIR spectral analysis

The FTIR spectrum of the sample in the range of 400–4000 cm−1 is shown in Fig. 3. The stretching mode at 3422 cm−1 and a weak asymmetric peak at 1620 cm−1 are characteristic of the O–H stretching vibration, due to the absorption of water molecules during the sample preparation process32. The vibration absorption peak at 1380 cm−1 corresponds to COO–33. Additionally, Cob exhibits stretching vibration bands of O–OH bonds at 3630 cm−1 and Co–O bonds at 499 cm−1, which are characteristic of Co(OH)234. With the increase in Co content in Fe, strong Co–O stretching and bending modes appear at 570 cm−1 and 663 cm−1, indicating the formation of Co3O4. The cubic structure of Co3O4 exhibits phase purity with a monodisperse nature35.This suggests that FCO-5, FCO-3, and FCO-1 have good stability and the Co3O4 produced by them has uniform particle size, consistent with the previous BET analysis results.

Fig. 3.

Fig. 3

Open in a new tab

FTIR spectra of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

XPS

In order to further investigate the impact of Co doping on FCO catalyst, the surface elemental composition and chemical state of FO, FCO-5, and Cob samples were analyzed using X-ray photoelectron spectroscopy (XPS). Figure 4a displays the Fe 2p spectra of FO and FCO-5. The Fe 2p3/2 and Fe 2p1/2 main signals of FO and FCO-5 can be observed at 710.0 eV and 723.4 eV, but the main peak of Fe 2p in FCO-5 is wider, and the satellite peak is smaller compared to FO. This indicates that the addition of Co promotes the formation of Fe3O4 in FCO-536, consistent with XRD results. The Fe 2p3/2 peak at 710.0 eV for FO and FCO-5 can be further decomposed into two components: Fe3+ (711 eV) and Fe2+ (709.6 eV). The Fe 2p1/2 peak at 723.4 eV for FO and FCO-5 can be resolved into two peaks: Fe3+ (725.0 eV) and Fe2+ (723.2 eV). In addition, the peaks at 717.5 eV and 732.4 eV are attributed to the vibration satellites of iron37. The addition of Co can regulate the chemical properties and distribution of grain boundary phases in FCO, thereby altering its magnetic properties38. This regulation is conducive to catalyzing the conversion of o-H2 to p- H2.

Fig. 4.

Fig. 4

Open in a new tab

XPS spectra of the (a) Fe 2p: FO and FCO-5 and (b) Co 2p: FCO-5 and Cob (sat, satellite).

Figure 4b illustrates the Co 2p spectra of FCO-5 and Cob. The signals observed at 779.1 eV and 794.2 eV in FCO-5 correspond to Co 2p3/2 and Co 2p1/2, respectively, with satellite peaks at 786.9 eV and 802.7 eV, suggesting the existence of Co3O439. The fitted peaks at 779.0 eV and 794.0 eV are attributed to Co3+, while those at 780.2 eV and 795.7 eV are assigned to Co2+. In the spectrum of Cob, peaks are identified at 780.2 eV and 795.7 eV for Co 2p3/2 and Co 2p1/2, respectively, with an orbital separation of ~ 15.5 eV, confirming the presence of Co2+40. Furthermore, the satellite peaks at 785.4 eV and 802.1 eV support the presence of the Co(OH)2 phase41, in agreement with XRD findings.

VSM

The magnetic properties of FO, FCO-5, and Cob were analyzed at 77 K by varying the magnetic field, disclosing distinct magnetic characteristics (Fig. 5). FO exhibits negligible coercivity and remanence values, displaying an S-shaped hysteresis loop indicative of superparamagnetic traits17. Despite reaching 90,000 Oe, the magnetization intensity of FO did not attain saturation, suggesting the presence of a spin-disordered region that remains active at low temperatures, resulting in a magnetization intensity of 14.12 emu·g−1. The hysteresis loop of Cob shows a straight line, with magnetization (M) proportional to magnetic field strength (H), indicating the material exhibits paramagnetism42. The hysteresis loop of FCO-5 exhibits a symmetric magnetic hysteresis curve, with a coercivity (Hc) of 538.08 Oe and a remanent magnetization of 7.29 emu·g−1, indicating that FCO-5 is a hard magnetic material. This is attributed to the presence of materials like CoFe2O4 in FCO-543. Taking CoFe2O4 as an example, the Co and Fe ions in CoFe2O4 occupy tetrahedral and octahedral sites, respectively. This unique arrangement allows for the mutual reinforcement of the magnetic moments of Co2+ and Fe3+ ions, resulting in a significant net magnetic moment44.

Fig. 5.

Fig. 5

Open in a new tab

Hysteresis loops of FO, FCO-5 and Cob at 77 K.

Catalyst activity testing

The catalyst samples were tested for activity at 77 K, and the results are shown in Fig. 6. FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 all exhibited good catalytic activity. The catalytic activity of Cob was significantly lower than that of the Fe-containing catalysts. As the Co doping content increased, the conversion rate and the content of outlet p-H2 of the FCO series catalysts initially increased, then decreased. The results indicated that FCO-5 had the best catalytic performance. The O-P conversion rate of Cob remained below 75%, and the content of outlet p-H2 after conversion remained below 40%. Under conditions of GHSV (gaseous hourly space velocity) at 1800 h−1, the normal para-hydrogen conversion rates of FO and FCO-1 were similar, both around 90%. But the content of outlet p-H2 after conversion of FCO-1 (48%) was higher than FO (41%). FCO-9, FCO-7, FCO-5, and FCO-3 all had O-P conversion rates exceeding 97%, with outlet para-hydrogen content above 49%. With the increase of GHSV, the catalytic activity of FCO-9, FCO-1, and Cob gradually decreased. However, the catalytic activity of FO increased with the increase of GHSV. When GHSV = 5400 h−1, the conversion rate of FO was close to FCO-9, reaching 95%, but the outlet p-H2 content of FCO-9 was higher than that of FO. Therefore, it can be concluded that the catalytic performance of Fe-based catalyst doped with Co is superior to the single-metal FO catalyst, and far higher than Cob catalyst. The increase of GHSV did not have a significant impact on the activity of FCO-7, FCO-5, and FCO-3, as they all maintained good catalytic activity. FCO-5 in particular exhibited stable catalytic activity, with a conversion rate of over 99% and an outlet p-H2 content levels above 49.6%, approaching the equilibrium concentration of p-H2 (~ 50%) at 77 K45.

Fig. 6.

Fig. 6

Open in a new tab

(a) O-P conversion and (b) content of p-H2 in the outlet after catalytic reaction of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob.

Table 3 shows the O-P conversion rate constants (K) of the sample catalysts. The highest O-P conversion rate constant was observed in FCO-5, GHSV = 5400 h−1, reaching 291.7 mol·L−1·s−1. The O-P conversion rate constants of FO and Cob were lower than the FCO series at all GHSV, indicating that the doping of Co improved the performance of Fe-based catalysts. FCO-7, FCO-5, and FCO-3 all showed a trend of increasing reaction rate constants with increasing GHSV. However, the reaction rate constant of FCO-9 showed a trend of first increasing and then decreasing, with the maximum value occurring at GHSV = 3600 h−1, reaching 280 mol·L−1·s−1. The O-P conversion rate constant of Cob was much lower than that of the Fe-containing catalyst.

Table 3.

The reaction rate constant K (mol·L−1·s−1) of FO, FCO-9, FCO-7, FCO-5, FCO-3, FCO-1 and Cob at 77 K.

GHSV (h−1) K(FO) K (FCO-9) K (FCO-7) K (FCO-5) K (FCO-3) K (FCO-1) K(Cob)
1800 20.5 134.2 97.3 98.1 81.2 52.3 13.2
3600 31.6 280.7 235.4 211.3 235.8 115.7 29.7
5400 86.5 180.2 255.0 291.7 253.5 107.9 15.8
Open in a new tab

Based on the experimental results above, the catalytic activity of the Fe-Co bimetallic catalysts in the FCO series remains consistently high. Among them, FCO-5 exhibits the highest reaction rate constant of 291.7 mol·L−1·s−1 among Fe-Co catalysts. Moreover, FCO-5 demonstrates exceptional stability with conversion rates consistently above 99% and outlet p-H2 content consistently above 49.6%. Therefore, the optimal Fe/Co ratio for the Fe-Co bimetallic catalysts is determined to be 1:1. Xu et al.25 prepared the FeCo bimetallic catalyst CFO-3 using NH4OH as a precipitant. At 77 K and a hydrogen flow rate of 100 mL·min−1, the p-H2 content was 47.61%. A comparison revealed that the catalysts prepared by Xu et al. exhibited entirely different patterns in XRD characterization with varying Co/Fe ratios compared to those in this study, demonstrating different crystalline phases due to the use of different precipitants. Additionally, Xu et al. suggest that the addition of Co increases the proportion of Fe3+, enhancing the catalyst’s magnetic strength and thus improving its catalytic performance. In contrast, this study proposes that the complex crystalline phases introduced by Co, which contribute to internal disorder and magnetic moments, are crucial factors in the excellent catalytic performance of FeCo bimetallic catalysts, rather than simply the enhanced magnetic characteristics of Fe3+ over Fe2+. Additionally, Xu et al.46 studied the ortho-para hydrogen conversion effects of different metal dopants and prepared the FeMn bimetallic catalyst Mn-FO, which achieved a p-H2 content of 49.45% after catalytic conversion at 77 K with a hydrogen flow rate of 100 mL·min−1. The p-H2 content at the outlet of the FCO-5 catalyst prepared in this study surpasses that of the aforementioned catalysts.

The doping level of Co significantly affects the catalytic performance of FCO. On one hand, varying amounts of Co doping result in different pore structures in FCO. Specifically, the addition of Co increases FCO’s lag coefficient and enhances surface porosity (Figure S1). The exposed pores and higher lag coefficient facilitate sufficient contact and interaction time between n-H2 and FCO, favoring ortho-para conversion. On the other hand, based on XRD (Figure S2) and FTIR (Figure S4) results of FCO-5 before and after the reaction, there is minimal difference, indicating no new phases, chemical bonds, or bond breaks occurred. This suggests a magnetic conversion process for o-H2 to p-H247. The addition of Co introduces new phases such as CoFe2O4, Co3O4, and Fe3O4 in FCO. Compared to single-component phases, this mixture increases internal disorder, leading to larger magnetic moments. Particularly in FCO-5, which contains components such as CoFe2O4, Co3O4, and Fe3O4, the arrangement of ions favors the generation of larger magnetic moments compared to Fe2O3 and Co(OH)2, thereby facilitating rapid o-p conversion.

Conclusion

In summary, the doped FCO catalyst series with Co elements were successfully prepared using a co-precipitation method. The study investigated the effect of the amount of Co element doping on the hydrogen conversion. The results indicate that doping with Co alters the pore structure of FCO, increasing the lag coefficient and thereby enhancing the contact time between the catalyst and n-H2. FCO-5 possesses a large specific surface area and the highest lag coefficient, making it the optimal catalyst for hydrogen conversion. Additionally, the incorporation of Co into FCO brings forth an increased number of active sites and larger magnetic moments, which facilitate the o-p conversion process. At a GHSV of 5400 h−1, the reaction rate constant for FCO-5 can reach 291.7 mol·L−1·s−1, and the p-H2 content can reach 49.7%. The preparation process of the FCO catalyst series is straightforward, the raw materials are inexpensive, and it has potential for industrial production and application.

Supplementary Information

Supplementary Information. (3MB, docx)

Acknowledgements

The authors acknowledge financial support from the National Natural Science Foundation of China (No. 51976094).

Author contributions

Conceptualization: L.Y., X.L., X.Z., methodology: K.S., Y.C., X.L., formal analysis: L.Y., X.Z., synthesis: L.Y., K.S., X.Z., writing—original draft preparation: X.Z., L.Y., writing—review and editing: L.Y., X.L. All authors have read and agreed to the published version of the manuscript.

Data availability

Our analyzed datasets are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-71790-9.

References

  • 1.Singh, A., Shivapuji, A. M. & Dasappa, S. VPSA process characterization for ISO quality green hydrogen generation using two practical multi-component biomass gasification feeds. Sep. Purif. Technol.315, 123667 (2023). 10.1016/j.seppur.2023.123667 [DOI] [Google Scholar]
  • 2.Li, Z., Zhang, W., Zhang, R. & Sun, H. Development of renewable energy multi-energy complementary hydrogen energy system (A Case Study in China): A review. Energy Explor. & Exploit.38, 2099–2127 (2020). 10.1177/0144598720953512 [DOI] [Google Scholar]
  • 3.Alkhaledi, A. N., Sampath, S. & Pilidis, P. A hydrogen fuelled LH2 tanker ship design. Ships Offshore Struct.17, 1555–1564 (2022). 10.1080/17445302.2021.1935626 [DOI] [Google Scholar]
  • 4.European Commission, A hydrogen strategy for a climate-neutral Europe https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52020DC0301 (2020).
  • 5.Office of Energy Efficiency and Renewable Energy, Hydrogen and Fuel Cell Technologies Office. Hydrogen Shot https://www.energy.gov/eere/fuelcells/hydrogen-shot (2021).
  • 6.Park, J. et al. 20 K H2 physisorption on metal-organic frameworks with enhanced dormancy compared to liquid hydrogen storage. ACS Appl. Energy Mate.6, 9057–9064 (2023). 10.1021/acsaem.2c01907 [DOI] [Google Scholar]
  • 7.Ratnakar, R. R. et al. Hydrogen supply chain and challenges in large-scale LH2 storage and transportation. Int. J. Hydrog. Energy46, 24149–24168 (2021). 10.1016/j.ijhydene.2021.05.025 [DOI] [Google Scholar]
  • 8.Zhou, H. et al. Study of activation methods for Ortho-para hydrogen catalysts in a small isothermal converter based on gas chromatography at LN2 temperature. Int. J. Hydrog. Energy55, 55–64 (2024). 10.1016/j.ijhydene.2023.10.155 [DOI] [Google Scholar]
  • 9.Lee, D. et al. Integrating hydrogen liquefaction with steam methane reforming and CO2 liquefaction processes using techno-economic perspectives. Energy Convers. Manage.245, 114620 (2021). 10.1016/j.enconman.2021.114620 [DOI] [Google Scholar]
  • 10.Lee, H., Haider, J., Abdul Qyyum, M., Choe, C. & Lim, H. An innovative high energy efficiency–based process enhancement of hydrogen liquefaction: Energy, exergy, and economic perspectives. Fuel320, 123964 (2022). 10.1016/j.fuel.2022.123964 [DOI] [Google Scholar]
  • 11.Aasadnia, M. & Mehrpooya, M. Large-scale liquid hydrogen production methods and approaches: A review. Appl. Energy212, 57–83 (2018). 10.1016/j.apenergy.2017.12.033 [DOI] [Google Scholar]
  • 12.Lv, C. et al. Design method of the hydrogen liquefaction process with catalyst-filled heat exchanger model. In Proceedings of the 10th hydrogen technology convention (eds Sun, H. et al.) 132–140 (Springer Nature Singapore, Singapore, 2024). [Google Scholar]
  • 13.Gyton, M. R., Royle, C. G., Beaumont, S. K., Duckett, S. B. & Weller, A. S. Mechanistic insights into molecular crystalline organometallic heterogeneous catalysis through parahydrogen-based nuclear magnetic resonance studies. J. Am. Chem. Soc.145, 2619–2629 (2023). 10.1021/jacs.2c12642 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Prisk, T., Bryan, M. & Sokol, P. Diffusive and rotational dynamics of condensed n-H-2 confined in MCM-41. Phys. Chem. Chem. Phys. PCCP.16, 17960 (2014). 10.1039/C4CP02281E [DOI] [PubMed] [Google Scholar]
  • 15.Teng, J. et al. Comparative study on thermodynamic performance of hydrogen liquefaction processes with various ortho-para hydrogen conversion methods. Energy271, 127016 (2023). 10.1016/j.energy.2023.127016 [DOI] [Google Scholar]
  • 16.Yang, J., Li, Y. & Tan, H. Integrated hydrogen liquefaction process with a dual-pressure organic Rankine cycle-assisted LNG regasification system: Design, comparison, and analysis. Appl. Energy347, 121372 (2023). 10.1016/j.apenergy.2023.121372 [DOI] [Google Scholar]
  • 17.You, H., Ahn, J., Jeong, S. & Chang, D. Effect of ortho-para conversion on economics of liquid hydrogen tanker with pressure cargo tanks. Ships Offshore Struct.13, 79–85 (2018). 10.1080/17445302.2018.1426526 [DOI] [Google Scholar]
  • 18.Minaev, B. F. & Aagren, H. Spin catalysis of Ortho-para hydrogen conversion. The J. of Phys. Chem.99, 8936–8940 (1995). 10.1021/j100021a071 [DOI] [Google Scholar]
  • 19.Petzinger, K. G. & Scalapino, D. J. Para- to Ortho-hydrogen conversion on magnetic surfaces. Phys. Rev. B8, 266–279 (1973). 10.1103/PhysRevB.8.266 [DOI] [Google Scholar]
  • 20.Bar, A. K., Pichon, C. & Sutter, J.-P. Magnetic anisotropy in two- to eight-coordinated transition–metal complexes: Recent developments in molecular magnetism. Coord. Chem. Rev.308, 346–380 (2016). 10.1016/j.ccr.2015.06.013 [DOI] [Google Scholar]
  • 21.Liu, J., Peng, C. & Shi, X. Preparation, characterization, and applications of Fe-based catalysts in advanced oxidation processes for organics removal: A review. Environ. Pollut.293, 118565 (2022). 10.1016/j.envpol.2021.118565 [DOI] [PubMed] [Google Scholar]
  • 22.Svadlenak, R. E. & Scott, A. B. The conversion of Ortho- to para hydrogen on iron oxide-zinc oxide catalysts1. J. Am. Chem. Soc.79, 5385–5388 (1957). 10.1021/ja01577a013 [DOI] [Google Scholar]
  • 23.Das, T., Kweon, S.-C., Choi, J.-G., Kim, S. Y. & Oh, I.-H. Spin conversion of hydrogen over LaFeO3/Al2O3 catalysts at low temperature: Synthesis, characterization and activity. Int. J. Hydrog. Energy40, 383–391 (2015). 10.1016/j.ijhydene.2014.10.137 [DOI] [Google Scholar]
  • 24.Karlsson, E. Catalytic ortho- to parahydrogen conversion in liquid hydrogen, https://lup.lub.lu.se/luur/download?fileOId=8929489&func=downloadFile&recordOId=8929488 (2017).
  • 25.Xu, H., Bi, S., Xue, M., Zhou, W. & Zhang, C. Amorphous cobalt iron oxide nanoparticles with high magnetization intensity for spin conversion of hydrogen at 77K. Int. J. Hydrog. Energy48, 31643–31652 (2023). 10.1016/j.ijhydene.2023.04.313 [DOI] [Google Scholar]
  • 26.Polyukhov, D. M. et al. Efficient MOF-Catalyzed Ortho-Para hydrogen conversion for practical liquefaction and energy storage. Acs Energy Lett.7, 4336–4341 (2022). 10.1021/acsenergylett.2c02149 [DOI] [Google Scholar]
  • 27.Fei, M. et al. Highly selective photocatalytic methane coupling by Au-modified Bi2WO6. ACS Catal.14, 1855–1861 (2024). 10.1021/acscatal.3c05534 [DOI] [Google Scholar]
  • 28.Deng, S. et al. Spatiotemporal control for integrated catalysis. Nat. Rev. Method. Prime.3, 28 (2023). 10.1038/s43586-023-00207-0 [DOI] [Google Scholar]
  • 29.Xu, Q.-Q. et al. Crystal phase determined Fe active sites on Fe2O3 (γ- and α-Fe2O3) yolk-shell microspheres and their phase dependent electrocatalytic oxygen evolution reaction. Appl. Surf. Sci.533, 147368 (2020). 10.1016/j.apsusc.2020.147368 [DOI] [Google Scholar]
  • 30.Zhang, F., Yan, H.-N., Jin, Y.-F., Zhai, L.-F. & Sun, M. Co-Fe synergy in CoxFe1-xWO4: The new type peroxymonosulfate activator for sulfamethoxazole degradation. Chem. Eng. J.461, 141989 (2023). 10.1016/j.cej.2023.141989 [DOI] [Google Scholar]
  • 31.Yu, H. et al. Dual-pore carbon shells for efficient removal of Humic acid from water. Chem. A Eur. J.23, 16249–16256 (2017). 10.1002/chem.201702318 [DOI] [PubMed] [Google Scholar]
  • 32.Athar, T., Hakeem, A., Topnani, N. & Hashmi, A. Wet synthesis of monodisperse cobalt oxide nanoparticles. Int. Sch. Res. Netw.2012, 691032 (2012). [Google Scholar]
  • 33.Taibi, M. et al. Layered nickel hydroxide salts: synthesis, characterization and magnetic behaviour in relation to the basal spacing. J. Mater. Chem.12, 3238–3244 (2002). 10.1039/B204087E [DOI] [Google Scholar]
  • 34.Wang, S., Wang, R., Chang, J., Hu, N. & Xu, C. Self-supporting Co3O4/Graphene hybrid films as binder-free anode materials for lithium ion batteries. Sci. Rep.8, 3182 (2018). 10.1038/s41598-018-21436-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ahmad, M. & Siddiqui, K. A. Synthesis of mixed ligand 3D cobalt MOF: Smart responsiveness towards photocatalytic dye degradation in environmental contaminants. J. Mol. Struct.1265, 133399 (2022). 10.1016/j.molstruc.2022.133399 [DOI] [Google Scholar]
  • 36.Tan, P. Active phase, catalytic activity, and induction period of Fe/zeolite material in nonoxidative aromatization of methane. J. Catal.338, 21–29 (2016). 10.1016/j.jcat.2016.01.027 [DOI] [Google Scholar]
  • 37.Mu, Y. et al. Highly selective electrochemical Baeyer-Villiger oxidation through oxygen atom transfer from water. J. Am. Chem. Soc.146, 13438–13444 (2024). 10.1021/jacs.4c02601 [DOI] [PubMed] [Google Scholar]
  • 38.Liang, Y., Deng, Q., Tan, X. H., Li, H. & Xu, H. The effects of Co on the enhancement of magnetic properties by modifying the intergranular phase in Nd-Fe-B alloys. Sci. Rep.9, 1758 (2019). 10.1038/s41598-018-36583-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Duan, Q. & Chen, H. Synthesis and electrochemical properties of Co3O4 nanoparticles by hydrothermal method at different temperatures. IOP Conf. Ser. Mater. Sci. Eng.207, 012020 (2018). 10.1088/1757-899X/207/1/012020 [DOI] [Google Scholar]
  • 40.Wang, L. et al. Construction of bundle-like cobalt/nickel hydroxide nanorods from metal organic framework for high-performance supercapacitors. J. Mater. Sci.: Mater. Electron.33, 10540–10550 (2022). [Google Scholar]
  • 41.Wang, J., Xie, T., Deng, Q., Wang, Y. & Liu, S. Three-dimensional interconnected Co(OH)2 nanosheets on Ti mesh as a highly sensitive electrochemical sensor for hydrazine detection. New J. Chem.43, 3218–3225 (2019). 10.1039/C8NJ06008H [DOI] [Google Scholar]
  • 42.Hilal, A. Magnetic components modeling including thermal effects for DC-DC converters virtual prototyping https://theses.hal.science/tel-01127289 (2014).
  • 43.Zhao, Z. et al. A general thermodynamics-triggered competitive growth model to guide the synthesis of two-dimensional nonlayered materials. Nat. Commun.14, 958 (2023). 10.1038/s41467-023-36619-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Chitu, L. et al. Structure and magnetic properties of CoFe2O4 and Fe3O4 nanoparticles. Mater. Sci. and Eng. C27, 1415–1417 (2007). 10.1016/j.msec.2006.07.036 [DOI] [Google Scholar]
  • 45.Nantogma, S., Joalland, B., Wilkens, K. & Chekmenev, E. Y. Clinical-scale production of nearly pure (>98.5%) parahydrogen and quantification by benchtop NMR spectroscopy. Anal. Chem.93, 3594–3601 (2021). 10.1021/acs.analchem.0c05129 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Xu, H., Wang, J., Han, Y., Xue, M. & Zhang, C. Effect of unpaired electron number elements (Al, Cr, Mn) doping in Fe2O3 on ortho to para hydrogen conversion at 77 K. J. Energy Storage74, 109512 (2023). 10.1016/j.est.2023.109512 [DOI] [Google Scholar]
  • 47.Kumada, T. et al. An ENDOR spectrum of H atoms in solid H2. Chem. Phys. Lett.288, 755–759 (1998). 10.1016/S0009-2614(98)00296-6 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information. (3MB, docx)

Data Availability Statement

Our analyzed datasets are available from the corresponding author on reasonable request.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES