Multimode-ultrasound and microwave assisted natural ternary deep eutectic solvent sequential pretreatments for corn straw biomass deconstruction under mild conditions.

Mild and effective pretreatments are essential to deconstruct lignocellulosic biomass so as to reuse cellulose content for value-added products. In this study, sequential multimode-ultrasound and microwave with natural ternary deep eutectic solvent (NATDES) pretreatments were used to deconstruct corn straw and optimized factors such as NATDES, ultrasonic, and microwave parameters. Results indicated that the ultrasound-NATDES or microwave-NATDES pretreatment could remove 37.86% and 52.36% lignin, respectively. When using sequential multimode-ultrasound and microwave assisted NATDES pretreatment, the delignification efficiency increased to 61.50%, and the cellulose content increased from 34.70% to 76.08%. In addition, the delignification of sequential multimode-ultrasound and microwave assisted NATDES pretreatment (under the mild conditions of microwave heating at 60 °C and 60 min) increased to 57.39%, and the cellulose content increased to 59.98%, too. This highlighted the effect of the combined ultrasound and microwave technology. Finally, the microstructural changes of mercury intrusion porosimeters, scanning electron microscopy, thermogravimetric, X-ray diffraction and Fourier transform mid-infrared spectroscopy were conducted to confirm the effectiveness of this method to deconstruct corn straw. A mechanism of the deconstruction of corn straw biomass in NATDES with the assistance of the sequential multimode-ultrasound and microwave heating was proposed. This research could open a window for future use of biomass energy by deconstructing lignocellulosic biomasses using environmentally friendly pretreatment methods.

Introduction

As the limited reserves of petrochemical energy are gradually depleted, biomass energy has been widely developed as a substitute [1], [2]. In order to utilize biomass resources, it is necessary to deconstruct them, which is separating cellulose, hemicellulose, and lignin and make it better for value addition [3], [4], [5]. Nowadays, common pretreatment methods for biomass include: ultrasound, microwave and chemical pretreatment [6], [7]. Each of them has different advantages and disadvantages [7]. For example, ultrasound treatment is a green method which does not add any chemical reagents, but the effect is very low. Microwave treatment on the other hand, can complete the delignification process quickly, but it consumes a lot of energy. While chemical treatment has higher efficiency and a short reflection time, but this usually causes environmental pollution. So, it is essential to develop a mild, green and efficient pretreatment method [8].

Ultrasound could affect the delignification process of biomass in two aspects [9], [10], [11]. Firstly, the ultrasound would destroy the surface structure which slightly removes the surface wax layer and silica of lignocellulose and reduce the biomass particle size [12]. Lastly, the ultrasound promotes the mixing of biomass and solvents, improves the mass transfer rate and changes the extractability of biomass, and ultimately the pretreatment efficiency is improved [13]. However, it is tough to realize the entire delignification process only by ultrasound. There is a certain heat treatment needed, too [13]. Especially in recent years, the widely used microwave heating could be efficiently applied to biomass structure. It can quickly cause molecular friction, so that the material can be heated uniformly as a whole to achieve the purpose of heating [14]. Another effective pretreatment method is chemical pretreatment, but its high cost and environmental protection are often questioned [15], [16]. Recently, the pretreatment of biomass with deep eutectic solvents (DES) has attracted great interest. This is mainly because it is cheap and green [17], [18]. They can be formed only by a simple combination of hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA) [19], [20], [21]. And those all formed by natural metabolites are defined as natural deep eutectic solvent (NADES).

Some works have shown that the acidic NADES solution is ideal for extracting hemicellulose and lignin. Kandanelli et al. [22] synthesized DES-OL system which removed about 50% of lignin, but this showed a low delignification efficiency. Then, some researchers synthesized a three-component deep eutectic solvent (3c-DES) and achieved breakthrough progress [23], [24], [25]. Especially the result of Xia et al. [22], which led to 98% lignin removal. Interestingly, although 3c-DES can separate most of the lignin in lignocellulosic biomass, the working environment of 3c-DES is usually harsh, such as high temperature and high pressure for a long time. Considering this, other scholars focused on shortening the reaction time with the aid of microwave [26], [27]. Nevertheless, the high temperature still creates a relatively large demand for equipment and the environment. Since then, Ong et al. [9] and Subhedar et al. [28] applied ultrasound to the pretreatment process and research the effect under the mild conditions. But it showed a decrease in the delignification. Therefore, how to balance the harshness of pretreatment conditions and the effectiveness of component separation is still a relatively big challenge.

Up till now, very limited open literature had reported on the effect of joint pretreatment to assist NADES. As far as we know, there is no article systematically pointing out the effect of ultrasonic, microwave and DES parameters in joint pretreatment on the delignification of corn straw. Based on this, DES with natural metabolites was synthesized and sequential multimode-ultrasound and microwave assisted DES pretreatment was used to deconstruct corn straw in this study. We firstly systematically studied the effects of ultrasonic, microwave and NATDES parameters during the pretreatment process and then evaluated the feasibility of separating corn straw components under the mild conditions. In addition, we attempted to explain the mechanism of joint pretreatment at the cellular level for the first time, which would fill up the related theoretical gaps.

Materials and methods

Raw materials and reagents

Corn straw was collected from a grocery in Zhenjiang (Jiangsu, China). The corn straw was mechanically pulverized, filtered through a 100-mesh sieve, and dried at 60 °C to a constant weight. The analytical reagents used in the experiment like oxalic acid (OA), glycerol (Gly), lactic acid (LA), aluminum chloride (AlCl3), ethanol (EA), sulfuric acid and choline chloride (ChCl) were purchased from the Huadongqihuabo company (Zhenjiang, Jiangsu, China).

Preparation of natural ternary deep eutectic solvent

The natural ternary deep eutectic solvents (NATDES) used in experiments were synthesized from three raw materials, obtained from a mixture of HBD and HBA in a specific molar ratios (ChCl: OA: EA is 2: 2: 1; ChCl: OA: LA is 2: 2: 1; ChCl: OA: Gly is 2: 2: 1; ChCl: Gly: AlCl3 is 2: 2: 0.66). Thereafter, the mixture was stirred at 500 rpm at 80 °C. Subsequently, the mixture was turned into a uniform and transparent solution, which was then stirred for 15 min. Finally, the prepared NATDES was put in a desiccator filled with CaCl2 to cool to room temperature and stored for use [22], [25].

Multimode-ultrasound and microwave assisted NATDES sequential pretreatment of biomass

The pretreatment of biomass was divided into two stages: the first stage was ultrasonic processing; the second stage was microwave processing.

Multimode-ultrasound as a first stage pretreatment

In the first stage, the sample and NATDES (1:10 ratio, w/v) in a test tube was pretreated using an ultrasonic bath reactor. The multimode-ultrasound device used in the experiment was equipped with three frequency generators (20, 40, and 60 kHz), which can be used in single, dual, or multi-frequency modes. The maximum power of the ultrasonic transmitter was 300 W, the volume of the sample tank was 6 L, and the three frequencies are evenly distributed at an angle 120° [29]. The model diagram of the ultrasound equipment is shown in Fig. S.1. Different modes refer to using one or more ultrasonic generators to work simultaneously and maintaining the same total power of the ultrasonic transmitter. The tubes containing samples were submerged to a depth of 4 cm, exactly in the center of the ultrasonic tank. Sonication was done at different frequency modes (single frequency, 20–60 kHz; dual frequency, 20 + 40, 20 + 60 and 40 + 60 kHz; tri-frequency, 20 + 40 + 60 kHz), different output powers 60, 120, 180, and 240 W (i.e. the power density 10, 20, 30, and 40 W/L), different ultrasonic pretreatment time durations (5, 10, 20, 30, 60, 90, 120, and 150 min), and the pulse on-time 20 s and off-time 10 s. A thermostat was used to keep the reactor water temperature at 25 ± 2 °C. A control was prepared by immersing the sample in water under the same processing conditions without applying ultrasonic waves.

The calorimetric power of ultrasound was measured according to the method of Mamvura et al. [30]. A measured amount of water was added to the ultrasonic reactor and allowed to be idled for a while. This is to enable the temperature of the pure water to be consistent with the ambient temperature. A Precise electronic digital thermometer was used to measure the temperature of pure water before and after the ultrasound pretreatment. The calorimetric power was calculated according to the following formula:where P is the calorimetric power; m is the quality of pure water; C is the specific-heat capacity of water (4.2 J/g K) at constant pressure; T is the temperature after ultrasound pretreatment; T is the temperature before ultrasound pretreatment; t is ultrasound time.

Table 1 showed the calorimetric power of ultrasonic equipment under different frequency and power modes. According to the data, the proportion of calorific power was less than 10% of the total power. It was a very small part of the value. Son et al. reported that small reactors had very little heat dissipation which could be ignored; large reactors had larger heat losses due to their larger contact area and volume. According to the ultrasonic model in Table 1 and Fig. S.1, the ultrasonic reactor we used was smaller and the calorimetric power was low. Therefore, we used input power to represent the power of ultrasound [31].

Table 1

The Ultrasonic calorimetric power in different power modes and frequency modes (W).

Ultrasonic frequency	Ultrasonic power
	60 W	120 W	180 W	240 W
20 kHz	4.20 ± 0.13	8.31 ± 0.08	14.33 ± 0.52	21.20 ± 1.71
40 kHz	5.95 ± 0.49	11.90 ± 0.99	15.40 ± 0.99	19.25 ± 0.50
60 kHz	5.25 ± 0.86	13.12 ± 0.74	20.65 ± 1.31	28.42 ± 0.87
20 + 40 kHz	6.82 ± 0.74	12.42 ± 0.25	17.15 ± 0.89	23.31 ± 0.30
20 + 60 kHz	8.47 ± 0.10	18.06 ± 0.62	23.59 ± 0.43	32.90 ± 1.00
40 + 60 kHz	6.30 ± 0.34	13.23 ± 1.04	22.19 ± 0.69	31.08 ± 0.59
20 + 40 + 60 kHz	12.60 ± 1.71	29.61 ± 0.62	35.70 ± 0.84	45.85 ± 0.48

Microwave heating as a second stage pretreatment

After the ultrasonic irradiation pretreatment, the samples were placed in a microwave (Hi-energy Corporation China, TANK-ECO) digestion apparatus for heat treatment. The samples were heated from room temperature to the specified temperature (60, 80, 100, 120 and 140 °C) at a rate of 15 °C/min for different time durations (30 s, 1 min, 5 min, 10 min, 20 min, 30 min, 40 min, 1 h and 2 h). When the reaction was complete, the samples were immediately cooled off to room temperature.

Recovery of biomass and lignin components

Pretreated mixtures each was added to 200 mL of acetone: water (50:50, v/v) and soaked overnight. Then the solid parts of the soaked samples were filtered under reduced pressure, dried in an oven at 60 °C, kept for subsequent analysis. Thereafter, the acetone in the filtrates was removed by vacuum rotary distillation, then an excess of anti-solvent water was added to the filtrates to precipitate lignin. The lignin solids were dried and stored for subsequent use.

Determination of the main components of biomass by NREL method

Corn straw composition analysis and procedure for identification of lignin were performed according to the acid hydrolysis standard laboratory analysis procedures of the National Renewable Energy Laboratory (NREL) [32]. The monosaccharide content was analyzed on a high-performance liquid chromatograph Agilent 1260 C equipped with a refractive index detector and a Bio-Rad Aminex HPS-87H column (7.8 × 300 mm). The column temperature was 60 °C. The mobile phase used was 5 mM H2SO4 min at a flow rate of 0.6 mL; the sample injection volume 20 μL.

The components of cellulose, hemicellulose and lignin were measured as described in the NREL method. The removal rate (L) of each component is based on the removal rate in the original sample and is related to the solid recovery rate. Then, L was calculated from the following formula:where S is the solid recovery rate; L is the removal rate of individual components; W is the mass of regenerated biomass; W is the mass of the original biomass; b is the content (%) of individual component in pretreated biomass; b is the content (%) of individual component in untreated biomass [27].

Characterization of biomass structure

Mercury intrusion porosimetry (MIP)

The Auto Pore IV 9510 mercury intrusion porosimeter was used to measure the specific surface area and pore structure of the sample. Before test, 30 mg sample was dried in an oven at 60 °C for 12 h. First, samples were put in a low-pressure porosimeter with pressure up to 0.2 psia (800 μm). Then, it moved in high pressure porosimeter with pressure 60,000 psia (3.6 nm). The equilibration time, both in the low and high pressure stage was 10 s. The surface tension of mercury was 0.458 N/m (485 dyn/cm), and the contact angle between the mercury and the pore surface was 130° [33].

Scanning electron microscopy analysis (SEM)

The microstructure of dried powder corn straw samples subjected to different pretreatment conditions was analyzed by SEM. The different pretreatment conditions included: untreated corn straw, only ultrasound and only microwave pretreatments, ultrasound and microwave sequential pretreatments under the intense condition (120 °C, 1 min), and the mild condition (60 °C, 60 min). Images were taken by using a tungsten filament a scanning electron microscope (JSM-7001F, JEOL, Tokyo, Japan) and the sample was ground into a powder and dried. After dealing with electricity powder, the change in microstructure was observed at an acceleration voltage 15 KV [34].

Thermogravimetric analysis (TGA)

The method described by Yu et al. [10] was adopted for the TGA analysis. A corn straw sample (5–10 mg) was put in an alumina crucible. The crucible was flushed with nitrogen gas (25 mL/min). Then, the sample was heated from 25 °C to 600 °C at a rate of 10 °C/min.

X-ray diffraction (XRD)

The crystallinity of corn straw under different pretreatment conditions was analyzed in a Cu target Kα-ray source XRD diffractometer at a maximum tube pressure of 40 kV and a maximum tube flow of 40 mA. The scanning range was 5–60° and the scanning speed was 5°/min. The step was 0.02°. The calculation of crystallinity was given by the following formula [10]:

I is the diffraction peak intensity of the 002 crystal plane, 2θ ≈ 22°, and I is the diffraction peak intensity of the amorphous region, 2θ ≈ 18°.

Fourier transform mid-infrared spectroscopy (FT-IR)

FT-IR analysis of corn straw samples was performed using a Rayleigh mid-infrared spectrometer. A tablet press was used to mix corn straw and potassium bromide together to form a transparent film. Scanning was done with a resolution of 8 cm−1. Each FTIR spectrum was an average of 28 scans at a wavelength of 400~4000 cm−1 [10].

Statistical analysis

All experimental data were obtained from three parallel experiments. The data were statistically analyzed by Origin8.6 software and expressed in the form of mean ± standard deviation. One-way analysis of variance (ANOVA) was used to test for significant differences between sample groups. Differences between means were made at 5% significance level.

Results and discussion

Deconstruction of corn straw by ultrasound and microwave assisted NATDES sequential pretreatment

Effects of different pretreatment approaches

The components of the original corn straw mainly include cellulose, hemicellulose and lignin as displayed in Table 2. These results were generally consistent with the reports of Chen et al. [27].

Table 2

The removal of cell wall components of corn straw pretreated with NATDES, ultrasound and different heating methods.

Pretreatment conditions	Solid recovery	Primary Cell Wall Composition (%)			Removal (%)
		Cellulose	Hemicellulose	Lignin	Cellulose	Hemicellulose	Lignin
Untreated	/	34.70 ± 2.99a	23.28 ± 2.08c	20.39 ± 0.39c	/	/	/
Ultrasound	76.42 ± 0.98	32.99 ± 0.33b	23.68 ± 0.76c	16.57 ± 0.52a	13.79 ± 0.98	22.27 ± 2.66	37.86 ± 2.00
HH-S	46.98 ± 1.42	59.47 ± 0.36c	7.29 ± 0.50b	17.81 ± 0.51a,b	19.50 ± 2.42	85.29 ± 0.99	58.91 ± 1.94
Microwave	43.22 ± 1.04	60.84 ± 0.71c	8.14 ± 1.03b	17.26 ± 0.49a,b	21.35 ± 5.58	89.53 ± 0.75	52.36 ± 0.84
U/S with HH-S	45.58 ± 3.00	59.87 ± 0.37c	5.34 ± 0.03a	21.34 ± 1.07c	24.22 ± 2.10	84.85 ± 2.20	63.36 ± 1.33
U/S with MW	37.93 ± 2.26	68.55 ± 1.27d	5.34 ± 0.07a	18.47 ± 0.92b	25.11 ± 3.34	91.30 ± 0.42	65.60 ± 2.59

U/S: Ultrasound pretreatment; HH-S: HH-S digital display constant temperature oil bath heating 100 °C for 240 min; MW: microwave heating 100 °C for 20 min.

Means (±SD) annotated with different letters in the same column are significantly different (p < 0.05).

Previous studies had shown that pretreatment could change the main components of biomass, especially cellulose content. In this study, an increase in the cellulose content under different pretreatment conditions was observed (Table 2). Ultrasound as the first stage of pretreatment, dislodged some lignin and hemicellulose by eroding and abrading the surface of the sample. However, the room temperature could not break the complex hydrogen bonding network formed between lignin, hemicellulose and cellulose. Moreover, the high viscous NATDES prevented the transmission of ultrasonic energy through the reaction medium, which eventually led to only a 12.80% increment in the cellulose content of biomass. According to previous reports, it was difficult to perform effective delignification at room temperature by only using ultrasound and NATDES pretreatment [9], [35]. This was mainly because of NATDES properties, which needed a certain operating/working temperature, and also the high viscosity of NATDES prevented the reaction to proceed smoothly. But using the ultrasound could improve the working efficiency of NATDES at a certain temperature or reduce the operating/working temperature.

The heating process as the second stage treatment is the main factor affecting the deconstruction of biomass materials. It affects the biomass fractionation by enhancing the disintegration of lignin-carbohydrate complexes by NATDES [19]. Comparing with conventional heating methods, microwave irradiation can drive molecular motion and cause the entire substance to generate heat by molecular frictions from the inside to the outside. Hence, the heating efficiency is extremely high [36]. As shown in Table 2, microwave heating significantly shortened the reaction time (microwave 20 min, HH-S 240 min) and showed a greater advantage when combined with the first stage of sonication, as it increased the content of cellulose of the original corn straw by 97.54%. The content of cellulose of corn straw pretreated with the ultrasound/microwave reached 68.55 ± 1.27%, which was 7.71% higher than that obtained after the microwave pretreatment (p < 0.05). Under these pretreatment conditions, the removal rates of lignin and hemicellulose were maximum (65.60 ± 2.59% and 91.30 ± 0.42%, respectively). The cellulose content of the sample pretreated with ultrasound/microwave was significantly higher by 12.80% and 72.52%, respectively, than that of the sample pretreated with ultrasound and the sample pretreated with microwave heating (p < 0.05), indicating that ultrasound and microwave combined pretreatment showed complementary effects. However, it was worth noticing that the lignin content of the corn straw after ultrasound/HH-S heating was 21.34 ± 1.07%. This value was statistically similar to the lignin content of the untreated sample (20.39 ± 0.39%). The reason for this opposite trend might be because of the long heating time, which caused the aggregation of lignin particles and accordingly led to a decrease in the efficiency of NATDES binding. Consequently, hemicellulose and part of lignin were removed and cellulose was retained, so the lignin content increased [9].

Effect of NATDES types and ultrasound parameters

The purpose of this study was to investigate the effects of NATDES types and ultrasonic radiation conditions on the separation of corn straw components. The selected heating condition was the microwave heating, and the reaction temperature was raised from room temperature to 120 °C for 20 min at a rate of 15 °C/min. Due to the different recalcitrance of the biomass, the effects of NATDES types on different types of biomass sources are quite different. In this study, four different types of NATDES were designed, and the results of the performance of these NATDESs on the disintegration of the corn straw cell wall components are shown in Fig. 1a. Among these NATDESs, ChCl/OA/Gly was the best in terms of the deconstruction of the cell wall components of corn straw, especially the cellulose and hemicellulose contents; the first component increased to 71.28 ± 3.96%, the other component reduced to 6.04 ± 1.36%. Similarly, the content of hemicellulose in the sample pretreated with ChCl/OA/EA reduced to 6.16 ± 0.98%, but more cellulose was lost in this sample. Moreover, all four NATDESs showed a similar ability to remove lignin. What is more, due to the large removal of hemicellulose and other components, the ChCl/OA/Gly group showed the highest cellulose content than the other treated groups (Fig. 1a). Comparing with the results of Kandanelli et al. [22] and Chen et al. [27], the content of cellulose had been further increased. What is more important, the NATDES that are composed of choline chloride, oxalic acid, and glycerin contain in its structure pure natural compounds; these compounds can be degraded and reused.

Fig. 1

Biomass component contents of pretreated corn straw: (a) type of NATDESs; (b) ultrasound time; (c) ultrasound power; (d) ultrasound frequency. (Control 1: primary corn straw; Control 2: corn straw with microwave heating; ChCl: Choline chloride; OA: Oxalic acid; EA: Ethanol; LA: Lactic acid; Gly: Glycerin; and all the heating conditions are 100 °C for 20 min).

In previous researches, ultrasound fractionated more lignin and hemicellulose components through destroying the surface of corn straw and increasing the accessibility of lignocellulose [10]. By prolonging the ultrasound time, the cellulose content of the corn straw gradually increased and the content of lignin and hemicellulose generally decreased (Fig. 1b). After 30 min of ultrasonic irradiation, the cellulose content reached the highest level (66.31 ± 0.90% of the total biomass) and the content of lignin and hemicellulose decreased to 17.64 ± 0.36% and 2.69 ± 0.65%, respectively. However, the cellulose content did not change significantly when the ultrasonic time was extended more than 30 min, but the lignin content increased slightly (Fig. 1b). Moreover, the effect of the ultrasound power on the corn straw components showed similar trends as the time of increasing/decreasing component content before/after a power threshold. The removal of lignin from corn straw decreased to lower values as the ultrasound power increased to higher values (Fig. 1c). The best deconstruction of corn straw was obtained at 120 W (Cellulose 69.26 ± 1.09%, lignin 17.70 ± 0.55% and hemicellulose 5.48 ± 0.52%). According to Ong et al. [9], the reason for this phenomenon was probably due to the supersaturation effect of ultrasound which caused the particles to gather at the place where the ultrasound effect is strong and consequently the agglomerated particles reduced the ability of NATDES to deconstruct the cellulose-hemicellulose-lignin matrix.

The frequency mode of the ultrasonic wave is also one of the important factors. It changes the period of the sound wave and consequently the cavitation [37], [38]. The cavitation time is longer at lower frequencies; the cavitation efficiency is higher at higher frequencies; the energy loss reported on using a single-frequency ultrasound treatment is reduced by using multi-frequency ultrasound treatment [12]. Thus, it is particularly critical to select a suitable ultrasound frequency mode. From the data in Fig. 1d, the effect of single frequency ultrasound is worse than the combined one. Among them, 40 kHz perform best. It had 72.14 ± 1.14% content of cellulose, 6.43 ± 0.08% hemicellulose and 14.23 ± 0.20% lignin. But this was less effective than the combined frequency ultrasonic mode. When using 20 kHz (60 W) + 40 kHz (60 W) ultrasound mode increased the cellulose content of regenerated corn straw to a maximum of 75.17 ± 0.89% and the overall content of lignin and hemicellulose was also reduced to 15.05 ± 0.38% and 4.64 ± 0.08%, respectively (Fig. 1d). This ultrasound mode was judged to be the best mode for the deconstruction of corn straw biomass. In addition, we noticed that there was a gap between the effect of the 60 kHz ultrasonic frequency mode with the 20 kHz and 40 kHz ultrasonic frequency mode. This was also reflected in the combined ultrasonic frequency mode. So, the 20 kHz + 40 kHz mode was a preferred ultrasound mode.

Overall, the deconstruction of corn straw by using NATDES composed of ChCl, OA and Gly assisted by multimode-ultrasound and microwave (100 °C for 10 min) significantly increased the percentage content of cellulose (from 34.7 ± 2.99% to 75.17 ± 0.89%) and less cellulose loss. During this pretreatment process, more than 90% of hemicellulose and about 60% of lignin were removed and the cellulose loss rate was less than 15%, which was the main reason for the increase in cellulose content (Fig. 1). Therefore, using NATDES (ChCl, OA and Gly) assisted by 20 kHz (60 W) + 40 kHz (60 W) ultrasound mode for 30 min are the best conditions for the deconstruction of the corn straw.

Feasibility of microwave condition for deconstructing corn straw

The heating conditions are the main factors affecting the deconstruction of biomass by NATDESs. Fig. 2a showed the changes in cellulose content of corn straw deconstructed by NATDES at different microwave heating times and temperatures. The cellulose content reached its maximum value at 100 °C, 5 min; 120 °C, 1 min; 140 °C, 30 s, then the content decreased after prolonging the heating time. This decrease might be due to increasing the residence time of the biomass in the microwave digester, which tended to aggregate lignin and reduce the chance of fractionating lignin. Lower temperatures led to reduced efficiency and higher temperatures accelerated the effect of aggregation. Compared to 100 °C and 140 °C, the microwave heating at 120 °C led to the highest cellulose content (76.08 ± 0.79%), and to correspond lignin and hemicellulose removal efficiencies of 61.50 ± 2.45% and 90.32 ± 0.50%, respectively (Fig. 2a). In summary, 120 °C and 1 min were chosen as the best microwave heating parameters for the deconstruction process.

Fig. 2

Effect of microwave heating time and temperature on cellulose content of the pretreated corn straw: (a) intense conditions; (b) mild conditions. Other conditions: ultrasound 20 (60 W) + 40 (60 W) kHz for 30 min; NATDES (Choline chloride: Oxalic acid: Glycerin).

In terms of current research, the work of deconstructing biomass under low pressures and temperatures is still in its infancy stage. In this study, the feasibility of deconstructing corn straw biomass at low temperatures (60 °C and 80 °C) was tested (Fig. 2b). Judging from the results, the pretreatment temperature 80 °C was preferred, which was beneficial to the reaction. After microwave heating at 80 °C for 40 min, the cellulose content of corn straw increased from 34.70 ± 2.99% to 59.98 ± 0.19%, while after heating at 60 °C for 60 min the cellulose content was improved, reaching 54.14 ± 0.20%, and the removal rates of lignin and hemicellulose reached 46.44 ± 0.41% and 81.77 ± 0.06%, respectively. The purpose of separating the three components of corn straw had been initially achieved. As for the reaction time, just overextending the microwave heating over 30 min did not lead to a useful improvement, for example, prolonging the time to reaching 120 min increased the removal rates of hemicellulose and lignin by only 1.0% (Fig. 1b). The deconstruction of corn straw was also tested using a mild microwave heating of low temperature, short time regime (60 °C, 5 min), that is, mild heating conditions. From the results (Fig. 2b), the cellulose content increased to 45.01 ± 0.96%, with the removal rates of lignin and hemicellulose as 39.19 ± 0.77% and 37.22 ± 2.79%, respectively. Despite its insignificant effect, pretreating of the corn straw pretreated under the mild heating conditions had further improved the enhancing effects of cellulose content and the removal of hemicellulose and lignin, compared with the experimental results of Jablonsky et al. [31]. This offers a possibility for deconstructing lignocellulose biomass under the mild conditions.

Lignin recovery

As stated in sections 3.2.3 and 3.2.4, sequential multimode-ultrasound and microwave under the intense conditions removed 61.50 ± 2.45% of the lignin found in corn straw. So, recovery and purity of lignin are conducive to better value-added utilization of biomass. The results of the recovery and purity of the lignin of corn straw pretreated in NATDES aided with the ultrasound/microwave heating at different conditions are shown in Table 3. The higher the lignin removal efficiency, the higher the efficiency of lignin recovery. The corn straw sequentially treated by the ultrasound and microwave under the intense conditions showed the highest lignin recovery (70.97 ± 2.58%). Under these conditions, NATDES has shown the best effects on corn straw biomass as can be seen in dissolving most of the lignin and in improving purity percentages of the recovered lignin as well. In this context, the purity of the recovered lignin of the different pretreated samples was above 70%. As can be seen, the higher the lignin recovery, the higher the purity of the lignin (Table 3). The introduction of ultrasound and microwave not only improved the ability of NATDES to solubilize lignin and break lignin-carbohydrate complex (LCC) bonds, but also improved the purity of fractionated lignin. Therefore, this pretreatment method could achieve a good effect of deconstructing biomass.

Table 3

Statistics of pore size distribution, crystallinity and recovered lignin of the regenerated corn straw biomass.

Pretreatment conditions	Pore size distribution			Crystallinity (%)	Lignin recovery (%)	Lignin purity (%)
	Porosity (%)	Total hole area (m2/g)	Bet surface area (m2/g)
Untreated	73.90	61.59	230.00	17.24	/	/
Ultrasound	86.30	97.63	241.50	18.17	54.99 ± 0.95	71.27 ± 1.14
Microwave	81.30	68.88	162.30	14.21	67.49 ± 4.24	84.30 ± 0.77
U/S, MW with intense conditions	80.30	65.97	342.70	12.82	70.98 ± 2.58	86.46 ± 0.76
U/S, MW with mild conditions	82.50	89.21	256.50	20.27	55.03 ± 1.47	74.60 ± 3.32

U/S: Ultrasound pretreatment; MW: microwave heating; intense conditions: MW heating at 120 °C for 1 min; mild conditions: MW heating at 60 °C for 60 min.

Characterization of corn straw microstructure

Mercury injection porosimetry can analyze the change of pore size distribution and specific surface area of biomass before and after pretreatments. Except for the microwave heating, all the pretreatments increased the specific surface area of corn straw (Table 3), which might be as a result of the removal of lignin and hemicellulose with NATDES. The sample pretreated with the ultrasound and microwave (under the intense conditions) showed the highest specific surface area (342.7 m2/g) and the cellulose content (76.08 ± 0.79%). This might be due to the effect of ultrasound which promoted the mixing of the solvent. In the first step, ultrasound destroyed the outermost layer of corn straw, allowing NATDES to infiltrate, and in the second step, it mixed the solvents and samples, which ultimately led to an increase in the effect of delignification. Loow et al. [39] also reported the related mechanism of specific surface area and biomass deconstruction. In addition, the SEM images (Fig. 4) also confirmed this hypothesis. In this context, the formation of holes on the surface of corn straw by ultrasound could enhance the flow of NATDES towards the interiors of biomass samples and participate in the reaction; accordingly, the biomass was more thoroughly deconstructed (Table 3 and Fig. 3a and b). In short, all the changes in the porosity and specific surface area of the corn straw process proved the effects of ultrasound, microwave, and NATDES during the pretreatment, and their combination could lead to improved efficiency.

Fig. 4

SEM images: a/b corn straw, c/d ultrasound treated, e/f microwave treated, g/h ultrasound-microwave with intense conditions, i/j ultrasound-microwave with mild conditions. a, c, e, g, and i: ×100 magnification images; b, d, f, h, and j: ×1000 magnification images.

Fig. 3

Pore size distribution (PSD) of corn straw after pretreatment. U/S: ultrasound; MW: microwave; intense conditions: MW heating 120 °C for 1 min; mild conditions: MW heating 60 °C for 60 min.

Scanning electron microscopy (SEM) images visually provided morphological changes in the biomass of corn straw pretreated with NATDES assisted by ultrasound and/or microwave heating (Fig. 4). The surface structure of the raw corn straw was tight with smooth fiber lines that showed an obvious sense of layering (Fig. 4a and b). After ultrasonic irradiation, the surface of the sample was eroded, which was possibly happening due to the grown microbubbles generated by ultrasonic cavitation that damaged the S-layer (secondary wall layer) during bursting (Fig. 4c and d). When exposed to only the microwave pretreatment, except the surface was corroded, the sample was also peeled off and exposed multiple neatly arranged cellulose bundles which was a manifestation of the destruction (Fig. 4e and f). This also led to an increase in the morphological area and a decrease in crystallinity (Table 3). At the same time, the joint pretreatment showed a better result. Under the intense conditions, not only the lignin and hemicellulose were stripped, but also the crystalline cellulose and pores disappeared in a large amount. Only the parallel microfibers remained (Fig. 4g and h). A similar phenomenon also appeared under the mild conditions (Fig. 4i and j). This was due to the combination of ultrasound and microwave that played a complementary role, which could strip a large amount of lignin and cellulose. Thus, more cellulose was exposed.

The TG curve and DTG curve were drawn from the mass loss rate of lignocellulosic biomass and temperature change with time. Generally, the pyrolysis trend of lignocellulosic biomass can be divided into four stages and each stage corresponds to the decomposition of different substances [40], [41]. As shown in Fig. 5a, all the samples lost 5–8% weight at 50–125 °C, that is the first stage of water evaporation. At 180–320 °C, hemicellulose began to crack and reached the optimal pyrolysis temperature at 300 °C. At 300–400 °C, cellulose gradually degraded and reached the optimal pyrolysis temperature at 385 °C. The fourth stage of pyrolysis of lignin as the main component was above 400 °C [42]. It was worth noting that the sample subjected to multimode-ultrasound and microwave sequential pretreatments under the intense conditions had a high hemicellulose loss rate and low cellulose loss rate as shown in Table 2. The main reason for the thermochemical properties change of corn straw was the change of its surface structure and composition. After ultrasound and microwave combined pretreatment, most of the lignin and almost all hemicellulose in the corn straw were stripped and with holes formed by cavitation. The efficiency of pyrolysis was improved, which were also reflected in the DTG image.

Fig. 5

The change of structure of corn straw under different characterization methods: (a) TG; (b) DTG; (c) XRD; (d) FTIR. U/S: ultrasound; MW: microwave; intense conditions: MW heating 120 °C for 1 min; mild conditions: MW heating 60 °C for 60 min.

From the DTG curves (Fig. 5b), the original sample, the ultrasonically treated sample, and the microwave-treated sample showed a small broad peak found at a temperature range 250–320 °C, which is the broad peak of hemicellulose. Comparing with all the pretreatment methods, the ultrasound and microwave sequential treatment eliminated the broad peak, which was due to the effective removal of hemicellulose components by the combined pretreatment. Furthermore, the appearance of a larger narrow peak observed at the temperature range of 320–380 °C represented the pyrolysis of cellulose. This peak shifted to the right due to the removal of large amounts of hemicellulose and crystalline cellulose after pretreatments. Similarly, under ultrasound microwave intense conditions, the narrow pyrolysis peak of cellulose is characterized by the significantly highest peak value and the broadest peak area compared with the narrow peaks of other pretreated samples. This was probably ascribed to the swelling of the exposed amorphous structure of the cellulose and to the decrease in the crystal structure of the cellulose, eventually promoting heat transfer [40]. In conclusion, multimode-ultrasound and microwave sequential pretreatment had the largest pyrolysis rate of hemicellulose, the largest pyrolysis peak area of hemicellulose, and the most obvious degree of weight loss after pyrolysis. The deconstruction process of corn stover was more thorough which confirmed the superiority of ultrasound and combined treatment.

X-ray diffraction was used to assess the change in biomass crystallinity. As shown in Fig. 5c and Table 3, two peaks were perceived by all the samples in the range of 18.2° and 22.5° to obtain the intensity of the amorphous and crystalline regions [43]. The CrI value of the original corn straw was 17.24%, this value increased to 18.17% and 20.27% after ultrasound and ultrasound/microwave (under the mild conditions) pretreatments, respectively. This increase was mainly because of the removal of the amorphous lignin and hemicellulose which exposed more crystalline cellulose and therefore the CrI of the corn straw increased [39]. However, it was interesting to note that when microwave was used to heat at higher temperatures, a drop in biomass crystallinity was unexpectedly observed. This may be that under the intense conditions, not only a large amount of hemicellulose and lignin was removed, but also some part of the crystalline cellulose. Kumar et al. [44] also showed the same results. So, for the data of crystallinity, under the intense conditions, the crystallinity of regenerated corn straw decreased to 12.82%. Comparing to the result of single microwave pretreatment (14.21%), the crystallinity of corn straw decreased again. This confirmed the role of ultrasound in the joint pretreatment and showed a positive synergistic effect after combining with microwave. Therefore, the deconstruction of corn straw under ultrasound/microwave intense conditions pretreatment was more complete.

FT-IR was used to evaluate the changes of main functional groups of corn straw under different pretreatment conditions; the FT-IR spectra are presented in Fig. 5d. All the studied corn straw samples had similar peaks found within the wavenumber range 900 cm−1 and 2900 cm−1, which were mainly assigned to stretching vibrations of β-glycosidic bonds and –OH groups of cellulose molecules. The significant changes of absorption peaks at 1235 cm−1 and 1430 cm−1 corresponded to the stretching vibrations of the ether bond and C–C bond on the phenyl ring of lignin. More importantly, the absorption peak that appeared at 1650 cm−1 was due to the presence of water in cellulose molecules. The more the content of cellulose, the more severe the water molecule vibration. As can be seen in Fig. 5d, the regenerated corn straw after multimode-ultrasound and microwave sequential pretreatments (under the intense conditions) had the strongest vibration, which was due to the increase in cellulose content stated previously after treatment (Table 2 and Fig. 2). Besides, the significant decrease in the absorption peak at 1735 cm−1 represented the cleavage of C=O acetyl functional groups found in hemicellulose, which led to the removal of hemicellulose [45]. The changes of the functional groups confirmed that the corn straw had been deconstructed under the ultrasound/microwave intense pretreatment conditions.

Proposed mechanism for sequential pretreatment of corn straw

It was previously reported that the structure of corn straw fiber could be divided into three layers [2]: the primary wall layer (P), the middle rubber layer (ML), and the secondary wall layer (S) (Fig. 6). Located at the outermost is the S layer which is mainly composed of lignin and xylan. Because of a large amount of lignin, it gives the cell wall toughness and resilience. The ML layer is located between the P layer and the S layer. It often contains hemicellulose, pectin and glycoproteins to connect the S layer and the P layer. The innermost layer is the P layer, which distributes the ordered cellulose molecules and a small amount of lignin. These cellulose molecules are interconnected to form a network of microfibril structures and are embedded between the hemicellulose and pectin of the ML layer [46].

Fig. 6

The proposed mechanism of deconstruction of corn straw biomass in NATDES with the assistance of the sequential multimode-ultrasound and microwave heating.

When corn straw is irradiated with ultrasound, microbubbles form on the fibre surface and in the water (Fig. 6) [38]. With the action of ultrasound, the asymmetry due to different cavitation effects at the interface and in pure liquid will let these microbubbles instantaneous collapse. This creates a “reclining force” on the flat solid surface and allows the microjets hit the interface. Finally, it forms faults or erosion effects on the surface and the secondary wall to fall off in a sheet or film form [34], [47]. Meanwhile, the NATDES penetrates the ML and P layers through faults or pores and combines with hemicellulose and lignin near the microfibrils (Fig. 6). The removal of lignin and hemicelluloses will expose more cellulose [10]. In addition, microwave as a heating method to overcome the heat generated by molecular motion occurs both inside and outside. So, when NATDES enters the P layer, with the internal heating, all parts of the entire sample are simultaneously deconstructed (Fig. 6). This will maximize separation efficiency and the cellulose content is further increased [48]. The results of MIP and SEM proved that the corn straw had been deconstructed (Figs. 3 and 4).

With time, the ultrasound gradually reaches saturation. Prolonging the time again does not produce better results and as the power of ultrasound increases, a “hot spot effect” occurs. Particles aggregate together and cause the surface area of the biomass to shrink, which ultimately reduces the dissolution of lignin by deep eutectic solvents [9]. The aforementioned results (Fig. 1b and c) discussed earlier in section 3.2.2 proved these assumptions.

Conclusions

By studying the changes of corn straw component contents (cellulose, hemicellulose and lignin) before and after pretreatment, the synergistic effect of sequential multimode-ultrasound and microwave processing with NATDES and the feasibility of continued component separation under the mild conditions had been proven. Under the intense conditions of the combined effects of ultrasound (20 k (60 W) and 40 k (60 W) for 30 min), microwave heating (120 °C, 1 min), and NATDES (ChCl: OA: Gly), 61.50 ± 2.45% lignin and 90.32 ± 0.50% hemicellulose was separated and the cellulose content was increased to 76.08 ± 0.79%. Moreover, under the mild conditions of the combined effects of ultrasound (20 k (60 W) and 40 k (60 W) for 30 min), microwave heating (60 °C, 60 min) and NATDES (ChCl: OA: Gly), 39.19% of lignin was removed and the cellulose content increased to 45.01%. So, it can be concluded that sequential multimode-ultrasound and microwave could effectively improve the destructive effect of NATDES on corn straw biomass. To the moment, the removal and recovery of lignin by treating with NATDESs was a bit low. Hence, future work to develop NATDES which is suitable to cope with the studied combined system that maximized the deconstruction of biomasses is needed.

Conflicts of interest

There are no conflicts to declare.

CRediT authorship contribution statement

Dong Yan: Conceptualization, Methodology, Formal analysis, Software, Data curation, Writing - original draft, Writing - review & editing, Visualization. Qinghua Ji: Methodology, Software, Validation, Supervision. Xiaojie Yu: Software, Resources, Supervision, Project administration. Mo Li: Software, Validation, Formal analysis, Supervision. Olugbenga Abiola Fakayode: Writing - original draft, Writing - review & editing, Supervision. Abu ElGasim A. Yagoub: Writing - original draft. Li Chen: Supervision, Project administration. Cunshan Zhou: Conceptualization, Methodology, Validation, Investigation, Resources, Writing - review & editing, Project administration, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.