A novel Brönsted-Lewis acidic heteropoly organic-inorganic salt: preparation and catalysis for rosin dimerization.

A novel Brönsted-Lewis acidic heteropoly organic-inorganic salt has been prepared via the replacement of protons in neat phosphotungstic acid with both organic and metal cations. This hybrid catalyst, Sm0.33[TEAPS]2PW12O40, exhibited satisfactory performance in the dimerization of rosin to prepare polymerized rosin Under optimum conditions (15.0 g rosin and 5.0 g Sm0.33[TEAPS]2PW12O40 catalyst in 18.0 mL toluene at 90 °C for 10 h), a polymerized rosin product with a softening point of 120.1 °C was obtained. In addition, the Sm0.33[TEAPS]2PW12O40 catalyst maintains excellent catalytic performance over five recycles.

Background

Polymerized rosin has a higher softening point, lighter color, and better stability than rosin, and is harder to oxidize. It is a key ingredient in oil paints, printing ink, adhesives, perfume, and more (Cheng et al. 1996; Chen 1992). The industrial preparation of polymerized rosin, employing aqueous mineral acids such as H2SO4 or ZnCl2/HCl, suffers from various shortcomings, including corrosion, pollution, and difficult recovery. Some environmentally friendly catalysts, such as solid superacids (Luo and Wu 1999; Gao et al. 2007), have been used to realize the clean polymerization of rosin. However, despite their superior separation, solid superacids exhibit insufficient recycling performance due to their uneven and vulnerable active components.

Acid-functionalized ionic liquids, a class of catalyst with the advantages of both aqueous and solid acids, have been applied successfully in many acid-catalyzed reactions (Paun et al. 2008; Fang et al. 2008; Hoang et al. 2005), including the polymerization of rosin (Liu et al. 2005, 2008). Abietic-type resin acids in gum rosin, which contain conjugated double bonds, undergo polymerization readily, but fir-type resin acids, having more steric hindrance, cannot. Brönsted acids are generally considered more apt to promote the isomerization of fir-type resin acids towards abietic-type resin acids (Scheme 1), while Lewis acids favor the dipolymerization of abietic-type resin acids (Scheme 2) (Cheng et al. 1996). As a result, acid catalysts comprising both Brönsted and Lewis acidity would exhibit a more outstanding catalytic performance in the preparation of polymerized rosin. We previously synthesized (3-sulfonic acid)-propyl-3-methylimidazoliume (and triethylammonium) chlorozincates and demonstrated their good catalytic efficiency and recycling performance in the polymerization of rosin, which can be attributed to the Brönsted–Lewis acidity of the ionic liquids and liquid–liquid two-phase process (Liu et al. 2008, 2009). However, using these catalysts is inconvenient due to their lengthy synthetic cycles and high viscosity. Most importantly, the Lewis acidity of the metal chlorides in the anions would be difficult to maintain due to stability issues, similar to aluminum chloride acid salt ionic liquids.

Scheme 1

The isomerization of fir-type resin acids towards abietic-type resin acids

Scheme 2

The dipolymerization of abietic-type resin acids

For the past few years, types of heteropoly organic salt catalytic materials have called attention for their potential water tolerance, acidity and self-separation performance (Leng et al. 2009a, b, 2012; Li et al. 2011, 2014; Shimizu et al. 2009; Sun et al. 2012; Zhou et al. 2014). It has been found that heteropoly anions with high charge numbers in these materials lead to higher melting points than conventional ionic liquids (Yuan et al. 2014). Furthermore, based on the high charge numbers of heteropoly anions, sulfated organic cations with Brönsted acidity and metal cations with Lewis acidity can act together as counterions to heteropoly anions, establishing novel Brönsted–Lewis acidic heteropoly organic–inorganic salts (Yu ST 2013). Herein, we report a heteropoly organic–inorganic catalyst, Sm0.33[TEAPS]2PW12O40, with Brönsted–Lewis acidity, which has different performances for melting point, solubility and acidity with both heteropoly compounds and ionic liquids. Moreover, the dimerization of rosin catalyzed by Sm0.33[TEAPS]2PW12O40 as a solid acid has been carried out to achieve an environmentally friendly process for polymerized rosin.

Experimental

Materials and methods

Analytical grade H3PW12O40 was dried at 180 °C. All other chemicals were of analytical grade and used without further purification. The 1H-NMR spectra of the catalyst and intermediates were recorded with a 500 MHz Bruker spectrometer in D2O. FT-IR spectra for catalyst samples (the Py-IR sample was mixed with pyridine (2:1, v/v) for 24 h prior to measurement) on KBr discs were recorded on a Nicolet iS10 FT-IR instrument. Melting points were measured using a conventional method on an X-4 type micro melting point apparatus. TG analysis was performed with a NETZSCH-TG 209 F1 Libra instruments in dry N2 at a heating rate of 20 °C/min from 30 to 800 °C.

The acidity of the prepared catalysts was determined by potentiometric titration (Shi and Pan 2008; Vazquez et al. 2000). A mixture containing the sample (0.5 g) and acetonitrile (30 mL) was mixed at the stable potential before being titrated with n-C4H9NH2 solution (0.05 mol/L in acetonitrile). The initial and jump potential values were measured by a pH meter to identify the acid strength and total acid amount in catalyst samples.

Catalysts

Contrastive catalyst, [TEAPS]3PW12O40 and H2[TEAPS]PW12O40, were prepared according to the literature (Leng et al. 2009b). Analogously, equimolar triethylamine and 1,3-propanesultone (0.10 mol) were dissolved in 80 mL ethyl acetate and stirred at 50 °C for 24 h under nitrogen atmosphere. The obtained white precipitate, 3-(triethylammonio)propane sulfonate, was filtered, washed with ethyl acetate and dried at 100 °C for 6 h. Next, a solution of intermediate 3-(triethylammonio)propane sulfonate (0.008 mol) and Sm(NO3)3·6H2O (0.0013 mol) in water was dropped into another aqueous solution of H3PW12O40 (0.004 mol). The mixture was stirred at room temperature for 24 h, distilled to remove water, and washed with ethyl acetate. Finally, the obtained Sm0.33[TEAPS]2PW12O40 solid was dried in a vacuum at 80 °C for 6 h (Leng et al. 2009b; Ramesh Kumar et al. 2012). Catalyst Sm0.66[TEAPS]PW12O40 was prepared by a similar method to that outlined above, using alternative materials proportion.

Dimerization of rosin

In batch experiments, heteropoly organic–inorganic salt catalyst (5.0 g) was added to a round-bottomed flask contained toluene (18.0 mL) and dissolved rosin (15.0 g). The resulting reaction mixture was stirred vigorously at 90 °C for 10 h and then cooled to room temperature. The solid catalyst was removed by centrifugation and directly reused without further treatment. The reaction solution, from which the toluene solvent had been separated, was distilled under low pressure (2 mmHg) at 260–270 °C (system temperature) and 180–210 °C (steam outlet temperature) for 30 min to remove low softening point materials, such as rosinol and some unpolymerized rosin, and obtain the polymerized rosin product. The ring and ball softening points of the products were determined by SYD-2806G numerical control asphalt softening point tester.

Results and discussion

Characterization of hybrid catalysts

Similar patterns were observed in the 1H-NMR (500 MHz, D2O) spectra of Sm0.33[TEAPS]2PW12O40, Sm0.66[TEAPS]PW12O40, [TEAPS]3PW12O40, H2[TEAPS]PW12O40 and their intermediate 3-(triethylammonio)propane sulfonate [δ 1.17 (t, 9H), 2.02 (m, 2H), 2.87 (t, 2H), 3.23 (m, 8H)], which supported the correct structure of the organic cations (Leng et al. 2009b). The FT-IR spectrum of Sm0.33[TEAPS]2PW12O40 is shown in Fig. 1, in comparison with those of reused Sm0.33[TEAPS]2PW12O40, H2[TEAPS]PW12O40, and neat H3PW12O40. It was observed that Sm0.33[TEAPS]2PW12O40, reused Sm0.33[TEAPS]2PW12O40, and H2[TEAPS]PW12O40 had four featured peaks similar to those of H3PW12O40 [1080 (P–O), 984 (W=O), 889 (W–Ob1–W) and 806 cm−l (W–Ob2–W)], which were assigned to the Keggin structure. Different degree shifts (around 1079, 981, 897 and 806 cm−l) and lower peak intensities compared to H3PW12O40 confirmed that the formation of new heteropoly organic–inorganic salts had been achieved without breaking the Keggin structure. In addition, featured peaks at 2987, 1485, 1394, 1127 and 1156 cm−1 of TEAPS+ (Liu et al. 2008; Leng et al. 2009b) also appeared in the spectra of the hybrid catalysts. The above results indicated that both TEAPS+ and PW12O403− were combined in all of the hybrid catalysts. Moreover, the Py-IR spectrum illustrated in Fig. 2 indicates the existence of both Brönsted acid sites (1538 cm−1) and Lewis acid sites (1456 cm−1) in Sm0.33[TEAPS]2PW12O40, which proved the presence of Brönsted–Lewis acidity derived from organic cations and Sm3+.

Fig. 1

FT-IR spectra for hybrid catalysts. a Sm0.33[TEAPS]2PW12O40. b Reused Sm0.33[TEAPS]2PW12O40. c H2[TEAPS]PW12O40. d H3PW12O40

Fig. 2

Py-IR spectra for Sm0.33[TEAPS]2PW12O40

Table 1 shows the basic properties of the designed Sm0.33[TEAPS]2PW12O40 catalyst and similar substances. It was found that the introduction of Sm3+ lead to samples having properties similar to inorganic salts (melting points >300 °C) and quite different to those of heteropoly ionic liquids [TEAPS]3PW12O40 and neat H3PW12O40. Concerning acidity, a lower acid strength and total acid amount was observed in hybrid samples with metal cations compared with [TEAPS]3PW12O40 and H3PW12O40. Besides, the poor solubility of Sm0.33[TEAPS]2PW12O40 in both water and organic solvent toluene forced it to be used as a solid catalyst in the prospective reaction.

Table 1

Properties of hybrid catalysts

Samples	Feature	Melting point (°C)	Solubility (25 °C)		Acid strength (mV)	Total acid amount (mmol/g)
			Water	Toluene
Sm0.33[TEAPS]2PW12O40	White powder	>300	Almost not	Not	679	0.428
Sm0.66[TEAPS]PW12O40	White powder	>300	Almost not	Not	676	0.148
[TEAPS]3PW12O40	White powder	178	Almost not	Not	718	0.873
H3PW12O40	Faint yellow/white crystal	95	Yes	Not	702	0.791

Figure 3 shows that Sm0.33[TEAPS]2PW12O40 loses unbound and crystal water (calculated weight loss: 2.66 %) between 30 and 200 °C. Moreover, the decomposition peaks of the organic cation matrix and sulfonic group were observed at 290 and 420 °C, respectively. The TG profile indicated that Sm0.33[TEAPS]2PW12O40 was quite stable below 200 °C.

Fig. 3

TG profiles of Sm0.33[TEAPS]2PW12O40

Catalytic performances of Sm0.33[TEAPS]2PW12O40 in the dimerization of rosin

Table 2 shows that a product with a similar softening point and acid value to gum rosin was obtained without catalyst. This was also observed in the reaction catalyzed by H3PW12O40, having high acid strength, indicating that Brönsted acidity derived from the antiprotons of H3PW12O40 had almost no effect on the promotion of rosin dimerization under the solid acid catalyst conditions. In addition, the Brönsted acidity derived from sulfonic acid groups in the organic cations alone did not have the necessary catalytic ability for dimerization, resulting in a product with 103.7 °C softening point. When both Sm3+ and organic cations containing sulfonic acid group [TEAPS]+ were co-introduced to the structure of a heteropoly compound, Brönsted–Lewis double acidity (Fig. 2) and, consequently, a high catalytic performance were achieved (Table 2). In particular, a favorable polymerized rosin product with a softening point of 120.1 °C [higher than the result of Reference Liu et al. (2009) but lower than that of reference Liu et al. (2008)] was obtained by Sm0.33[TEAPS]2PW12O40, a hybrid catalyst with proper Brönsted and Lewis acidity proportions. However, the Lewis acid sites of Sm3+ in Sm0.33[TEAPS]2PW12O40 were surrounded by more bulky phosphotungstic acid radicals and organic cations than those of Sm0.66[TEAPS]PW12O40 and the Brönsted–Lewis Acidic ionic liquids reported in References (Liu et al. 2008, 2009). For this reason, the decarboxylated conjugated resin acids with lower steric hindrance attached more easily to Sm3+ than the conjugated resin acids without decarboxylation, which accounted for the lower acid value of products obtained from the Sm0.33[TEAPS]2PW12O40 catalyst (Liu et al. 2008, 2009).

Table 2

Catalytic dimerization performance of hybrid catalysts

Catalysts	Softening point (°C)	Acid value (mg/g)
Gum rosin	80.4	164.0
Blank	88.3	165.0
H3PW12O40	86.5	137.8
[TEAPS]3PW12O40	103.7	134.7
Sm0.66[TEAPS]PW12O40	111.9	147.9
Sm0.33[TEAPS]2PW12O40	120.1	111.8

Reaction conditions: catalyst 5.0 g, rosin 15 g, toluene 18 mL, 90 °C, 10 h

The solid catalyst, Sm0.33[TEAPS]2PW12O40, precipitated to the bottom of the reactor after the reaction, allowing separation from the solution via simple centrifugation. The separated Sm0.33[TEAPS]2PW12O40 catalyst could be reused several times without treatment (see Table 3). The polymerized rosin products with high softening points (above 120 °C) were obtained even after reusing the catalyst five times, which was attributed to the concerted catalysis of Brönsted and Lewis acid sites and their covalent or ionic bonding pattern in the body of the catalyst in the solid–liquid catalytic system. The FT-IR spectrum of reused Sm0.33[TEAPS]2PW12O40, shown in Fig. 2, also indicates that no apparent structural change had taken place in the catalyst during use.

Table 3

Catalytic reusability of Sm0.33[TEAPS]2PW12O40 for the rosin dimerization

Catalyst recycle times	Softening point (°C)	Acid value (mg/g)
0	120.1	111.8
1	114.3	125.8
2	115.0	120.3
3	126.1	124.5
4	120.4	93.7
5	120.3	96.0

Reaction conditions: catalyst 5.0 g, rosin 15 g, toluene 18 mL, 90 °C, 10 h

Conclusion

A novel heteropoly organic–inorganic salt with Brönsted–Lewis double acidity, Sm0.33[TEAPS]2PW12O40, was prepared via the replacement of protons in neat phosphotungstic acid with both organic cations containing sulfonic acid groups and metal Sm3+ cations. As a solid acid catalyst, this environmentally benign Brönsted–Lewis double acidic hybrid enables an effective catalytic performance in the dimerization of rosin to afford polymerized rosin products with a softening point above 115 °C. Moreover, the catalyst also exhibited reasonable reuseability, demonstrated by a five-run recycling test.