Introduction

The ability of carbon atoms to form covalent bonds with other atoms via sp, sp2 and sp3 enables it to form different allotropies such as, diamond, graphite, carbon nanotube, carbon dots and so on [1]. Carbon-based quantum dots with fascinating properties such as great potential applications in bioimaging, optoelectronics sensors, SERS, and photocatalyst have gradually become a rising star as a new nanocarbon member due to their benign, abundant and inexpensive nature [2,3,4]. A great variety of techniques and methods have been developed in recent years for preparation of C-dots such as chemical ablation [5,6,7], electrochemical carbonization [8,9,10,11], laser ablation [12], microwave irradiation [13] solvothermal and hydrothermal methods [14] are provided. On the other hand, other methods such as thermal decomposition of organic compounds in solvent with high boiling points is a simple and inexpensive procedure for preparing C-dots. This method is based on heating nitrogen-and oxygen rich organic molecules in an autoclave above 200 °C. Zhang et al. have reported N-doped C-dots can be formed by heating toluene solution of CCl4 and NaNH2 [15]. Other methods have also been used for preparation C-dots such as thermal decomposition [16, 17] and ultrasonic carbonization [18].

Recently a single step method based on thermal decomposition of substituted ammonium citrate salts are reported. Based on reported citric acid is one of the stating materials as carbon source in which solutions of branched poly(ethyleneimine) (BPEI) in water at 200 °C causes formation of C-dots modified with (BPEI) [16].

In this study, attempts have been made to use this type of C-dots as catalyst for Knoevenagel reactions because of basic character of C-dots surfaces.

The Knoevenagel reaction is due to the carbon–carbon bond formation between active methylene group and aldehydes or ketones. The Knoevenagel products have been widely used for the preparation of coumarin derivatives, cosmetics, perfumes, pharmaceutical compounds, polymers and so on [19,20,21]. Up to now various Lewis acids and bases have conventional been used as homogeneous catalyst [22,23,24,25,26]. Because of separation and recyclability problems of homogeneous catalyst, many efforts has been made to use heterogeneous system such as molecular sieves [27] organic functionalized molecular sieves, or silica [28,29,30] metal organic frameworks (MOF) such as UiO-66 [31], Zif-8 [32], Zif-9 [33], IRMOF [34], and mixed oxides nanoparticles [35] are active for Knoevenagel reactions.

Experimental

All chemicals were purchased from Merck Chemical Company, and used without further purification. XRD patterns of powdered samples were carried out on Siefert 3003 PTS diffractometer using Cu Kα radiation (k = 1.5406 Å) in the 2θ range of 10° to 60° with accelerating voltage and current of 40 kV and 40 mA, respectively. FT-IR spectra were recorded on a Bruker instrument using KBr pellets technique in the range of 4000–500 cm−1. The Atomic Force Microscopy (AFM) images were taken by Nanosurf easyscan 2. The products were analyzed by gas chromatography (GC) (Agilent Technologies, 6890 Series GC System, HP-5 Phenyl Methyl Siloxane Capillary, 30 m × 530 µm × 1.5 µm Nominal, Carrier Gas; He) GC–MS analyses were performed using an 5973 Network Mass Selective Detector, 6890 Network, GC System, Column; HP Phenyl Methyl Siloxane Capillary, 30 m × 530 µm × 1.5 µm Nominal, Carrier Gas; He. High Resolution Transmission Electron Microscopy (HRTEM) were performed using a Tecnai, FEI.

Preparation BPEI-CDs as catalyst

BPEI-CDs was prepared as reported [16,17,18]. CDs with BPEI coatings are produced by a mixture of citric acid (CA) with branched polyethylene imine (BPEI) at temperature (200 °C). Therefore 0.5 g of BPEI and 1 g of CA are dissolved uniformly with 10 mL of hot water in a 25 mL beaker, and then heated moderately (< 200 °C) using a heating mantle. Most of water evaporated for about 20 min, until a uniform pale-yellow gel was formed. Then one 1 mL of water was added before the gel was scorched and heating was continued. This procedure was repeated about 10 times (in 3 h), then the gel with orange color was obtained. The product was used without further purification. In this step, carbon dots with polyethyleneimine was obtained.

Catalyst test

In a typical experiment BPEI-CDs (25 mg), malononitrile (5 mmol), aldehyde (5 mmol) and solvent as ethanol (1.5 mL) were placed into round bottom flask (25 mL) and the reaction mixture was stirred at 60 °C for 2 h. Then the mixture was dissolved in 2 mL ethyl acetate and the Knoevenagel products were separated and subjected to GC and GC–MS analysis. The catalyst was recovered by simple filtration, washing with ethanol and drying in air in order to use in the next run.

Result and discussion

Catalyst characterization

In this study, BPEI-CDs was prepared based on reported [16], with citric acid and branched polyethylene imine.

The XRD pattern of BPEI-CDs is shown in Fig. 1, the obtained results is consistent with those reported before [16]. The typical diffractions at 2θ = 21.68° and 43° are related to the (002) and (100) planes respectively. In fact, the obtained results of BPEI-CD indicates the presence of carbon-based materials and graphite crystalline plates [16, 36].

Fig. 1
figure 1

The XRD pattern of BPEI-CDs

Full size image

The FT-IR spectra of citric acid, BPEI and BPEI-CD are shown in Fig. 2a–c respectively. The obtained results were consistent with those reported before [16]. The FT-IR spectrum of the BPEI–CDs is similar to that of the BPEI (3440 and 1585 cm−1 due to NH CH2 at 2950 and 2820 cm−1 and CN at 1338 cm−1), but no characteristic absorption of CA was observed. These results indicate that CA might be mostly carbonized during the pyrolysis, while the BPEI kept stable. In fact, observing a peak at 1700 cm−1 in Fig. 2c should be attributed to the –CONH stretching vibrations.

Fig. 2
figure 2

FT-IR spectra of (a) citric acid, (b) BPEI, (c) BPEI-CDs

Full size image

The AFM images of BPEI-CDs are shown in Fig. 3 the particle size distribution of BPEI-CD is 5 to 50 nm. Most particles have a size of 5 to 10 nm.

Fig. 3
figure 3

AFM imag of BPEI-CDs with two maps

Full size image

The HRTEM with two different magnification of modified C-dots are shown in Fig. 4. HRTEM images (Figs. 4a and 4b) reveals that BPEI-CDs spherical morphology with average particle size of 5–10 nm, but the majority of particle sizes are 5 nm. Based on the SAED results, the spacing between the crystalline lattice is around 0.26 nm.

Fig. 4
figure 4

HRTM image of BPEI-CDs with two magnifications

Full size image

The EDX of CDs is shown in Fig. 5, which indicates the presence of carbon, nitrogen and oxygen in the prepared BPEI-CDs.

Fig. 5
figure 5

The EDX of BPEI-CDs

Full size image

Catalytic activity

Optimization of the Knoevenagel condensation reaction

The Knoevenagel condensation reaction of benzaldehyde with malononitrile was performed in the presence of BPEI-CD as catalyst to form benzylidene malononitrile (Scheme 1).

Scheme 1
scheme 1

The Knoevenagel condensation reaction of benzaldehyde derivatives with malononitrile

Full size image

Various parameters such as the effect of catalyst amount, temperature, reaction time and solvent were evaluated. At first, to optimize the catalyst amount, reaction was carried out using 5, 13, 17 and 25 mg of BPEI-CD as catalyst within 2 h, in refluxing ethanol (Fig. 6). As seen in this figure, whereas increasing the amount of catalyst from 5 to 25 mg within 2 h increases the conversion from 87 to 100% with 72 to 100% selectivity toward the benzylidene malononitrile (Fig. 6).

Fig. 6
figure 6

Influence of catalyst amount on Knoevenagel condensation. Reaction conditions: benzaldehyde (5 mmol), malononitrile (5 mmol), solvent: ethanol (1.5 ml), 60 °C, 2 h

Full size image

Then, the effects of reaction temperature on catalytic activity were investigated (Fig. 7). The results showed that the reaction conversion gradually increased from 30 °C to 50 °C, but the maximum conversion (100%) of the product was obtained at 60 °C within 2 h.

Fig. 7
figure 7

Influence of reaction temperature on Knoevenagel condensation. Reaction conditions: benzaldehyde (5 mmol), malononitrile (5 mmol), solvent: ethanol (1.5 ml), catalyst 0.025 g, 2 h

Full size image

In order to evaluate the effect of time on the Knoevenagel reaction, the reaction was performed at 60 °C for different reaction times from 30 min to 2 h (Fig. 8). With increasing reaction time, the conversion of benzaldehyde increased to 100% after 2 h. Therefore, all other reactions were carried out within 2 h.

Fig. 8
figure 8

Influence of reaction time on Knoevenagel condensation. Reaction conditions: benzaldehyde (5 mmol), malononitrile (5 mmol), solvent: ethanol (1.5 ml), catalyst 0.025 g, 60 °C

Full size image

In order to investigate the effect of solvent on the product formation, reactions were carried out with different solvents such as methanol, ethanol, butanol, DMF and toluene (Fig. 9). It was found that the highest and lowest catalytic activity is observed in ethanol and toluene, respectively. As such, it can be concluded that the less polar solvents such as toluene are not suitable media for this reaction. In contrast, ethanol as a rather polar and eco-friendly solvent is better perhaps due to the stabilization of the generated partial ionic transition state structure. In fact, the reactivity of the catalyzed reactions to solvent depends on the (a) polarity and amphiprotic properties of catalyst surface [32, 33, 37]. For example, whereas some reactions in the presence of MOF catalyst such as UiO-66-NH2 exhibits the highest catalytic activity in DMF, ethanol is a good media as well. On the other hand, ethanol was found to be better than DMF in cases such IRMOF-3 [37] and NH2-MIL(Al) [32] catalysts. This may be is due to the amphiprotic character of solid surfaces [34]. It was also reported that toluene is better than polar solvents for some amino-tagged silicas [33, 38]. In contrast, the polar solvent increases the reaction rate in some other Knoevenagel catalyzed reactions. Therefore, there is no general trend in the effect of solvent on Knoevenagel reactions using heterogeneous catalysis system.

Fig. 9
figure 9

The effect of solvent on product formation reaction conditions: aldehyde (5 mmol), malononitrile (5 mmol), catalyst 0.025 g, 60 °C, 2 h

Full size image

Having established the optimal reaction conditions, we examined the generality of this Knoevenagel condensation on other substrates such as 3-nitrobenzaldehyde, 2-chlorobenzaldehyde, 4-chlorobenzaldehyde, 2-hydroxybenzaldehyde, 4-hydroxybenzaldehyde, 4-methylbenzaldehyde, 4-methoxybenzaldehyde and furfural in ethanol (1.5 mL) using 0.025 g of catalyst within 2 h, at 60 °C (Table 1).

Table 1 Results obtained for Knoevenagel condensation of aldehydes
Full size table

Effect of the substitution on the conversion percentages can be rationalized if one has an insight into the reaction mechanism. As indicated in Scheme 2, it seems likely that the BPEI-CDS as Lewis base (LB) abstracts a H+ from malononitrile and generates compound A. Subsequently, A reacts with aldehyde to afford B. Protonation followed by elimination of water results in the formation of C and Knoevenagel adduct D, respectively. Effect of the substitutions on the conversion rate supports the suggested mechanism. As seen in the Scheme 2, the presence of electron-withdrawing group in meta or para position of aldehyde should accelerate the reaction rate due to the stabilization of the intermediate B. This in turn increases the conversion rate (entries 1, 2 and 4, Table 1). On the other hand, the presence of electron-donating groups on para position retards the conversion rate due to the destabilization of the intermediate B. This in turn gives the corresponding product in lower yield (entries 6, 7 and 8, Table 1). That ortho-chloroaldehyde has afforded the product in lower yield (entry 3, Table 1) mas has been the result of partial steric effect experienced between the rather large chlorine atom and reaction center. Finally, observation of excellent yield in the case of ortho-hydroxy aldehyde (entry 5, Table 1) is not surprising since the formation of hydrogen bonding between the –OH group and –C–O− in the transition state increases the conversion rate (inset, Scheme 2, E).

Scheme 2
scheme 2

Suggested mechanism for catalyzed Knoevenagel condensation reaction of aldehydes with malononitril. Inset, the stabilized intermediate E due to the formation of intramolecular hydrogen bond

Full size image

The recyclability of BPEI-CDs was investigated under optimum reaction conditions. After each reaction run the catalyst was recovered by centrifugation, washed with ethanol for several times and dried under vacuum at 100 °C. As shown in Fig. 10, the catalyst maintains with high catalytic activity (100% conversion and 100% selectivity), after three recycling step a very slight decreasing was observed (with 99% conversion and 98% selectivity). On the other hand, no catalytic activity was observed using the filtrate solution. The obtained results indicate the heterogeneous character of the prepared catalyst.

Fig. 10
figure 10

The effect of recyclability of catalyst for Knoevenagel condensation reaction at optimum conditions

Full size image

The other point is the FT-IR spectra before and after experiment were similar (Fig. 11).

Fig. 11
figure 11

FT-IR spectra of BPEI-CDs before and after Knoevenagel reaction

Full size image

The XRD patterns of BPEI-CDs before and after reaction is shown in Fig. S1. The intensity of broad diffraction peak at 2θ = 24.0 (d002) was decreased and the intensity of weak peak at 43.4 (d100) has been increased. These results indicate the amorphous carbon phase decreased and graphitic phase increased which are consistent with those reported before [36]. The interesting point is the XRD pattern of BPEI-CDs after second and third run were similar.

Finally, in order to compare our results with those reported in literature, Table 2 was included. The preparation of BPEI-CDs as a new Knoevenagel heterogeneous catalyst in one step in a free metal procedure together with low reaction time, and high conversion and selectivity are some advantages of this work.

Table 2 Comparison of present work (catalyst BPEI-capped CDs) with other studies in the literature
Full size table

Conclusion

In this work modified Carbon dots with citric acid and branched polyethylenimine (BPEI) was prepared and designated as BPEI-CDs. The prepared BPEI-CDs was characterized using with a variety of different techniques, including FT-IR, XRD, HRTEM and AFM. The BPEI-CDs was used as basic catalyst for Knoevenagel reactions with aromatic aldehydes and malononitrile. It was found that the unsubstituted aldehyde and substituted with electron-withdrawing groups are more active than those containing electron-releasing groups. It was also found that the catalyst can be easily recovered from reaction mixtures and reused at least three times without significant loss in catalytic activity.