Accepted Manuscript

Title: Homogenization-assisted cavitation hybrid rotation

extraction and macroporous resin enrichment of
dihydroquercetin from Larix gmelinii

Authors: Yu Xia, Yinhang Wang, Wei Li, Chunhui Ma,

Shouxin Liu

PII: S1570-0232(17)30737-7
DOI: https://doi.org/10.1016/j.jchromb.2017.10.044
Reference: CHROMB 20874

To appear in: Journal of Chromatography B

Received date: 24-4-2017

Revised date: 15-10-2017
Accepted date: 20-10-2017

Please cite this article as: Yu Xia, Yinhang Wang, Wei Li, Chunhui Ma, Shouxin
Liu, Homogenization-assisted cavitation hybrid rotation extraction and macroporous
resin enrichment of dihydroquercetin from Larix gmelinii, Journal of Chromatography
B https://doi.org/10.1016/j.jchromb.2017.10.044

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 Homogenization-assisted cavitation hybrid rotation extraction

and macroporous resin enrichment of dihydroquercetin from

Larix gmelinii
Yu Xia, Yinhang Wang, Wei Li, Chunhui Ma*, Shouxin Liu**

a
College of Material Science and Engineering, Northeast Forestry University, 150040 Harbin,

China

*, ** Corresponding College of Material Science and Engineering, Northeast Forestry

University, 150040 Harbin, China;

Tel.: +86-451-82191204, Fax: +86-451-82191204 (Chunhui Ma); Tel.:

+86-451-82191502, Fax: +86-451-82191502 (Shouxin Liu).

E-mail addresses: mchmchmchmch@163.com (Chunhui Ma), liushouxin@126.com (Shouxin

Liu).

Graphical abstract
Research Highlights
• Homogenate-assisted cavitation hybrid rotation extraction was applied for extracting DHQ;
• The mechanism of homogenate-assisted has been interpreted in detail;
• The mechanism of cavitation hybrid rotation extraction has been interpreted in detail;
• Adsorption behavior of DHQ on AB-8 resin column chromatography was described in detail.

Abstract

Cavitation hybrid rotation, which was and is still looked upon as an unavoidable nuisance in

the flow systems, for extraction processing intensification of active chemical compounds from

natural products. In this study, a homogenization-assisted cavitation hybrid rotation extraction

method was applied to extract dihydroquercetin (DHQ) from larch (Larix gmelinii) wood root.

The extraction parameters were optimized in single factor experiments with the DHQ

extraction yields as the response values. The optimum conditions were as follows: number of

extractions, three; ethanol volume fraction for the extraction, 60%; liquid–solid ratio for

homogenization, 10 mL/g; homogenization time, 8 min; liquid–solid ratio for cavitation

extraction, 9 mL/g, and cavitation extraction time, 35 min. Under these conditions, the DHQ

content in extract was 4.50±0.02 mg/g, and the extraction efficiency was higher than those of

traditional techniques. Cavitation can be effectively used to improve the extraction rate by

increasing the mass transfer rates and possible rupture of cell wall due to formation of

microcavities leading to higher product yields with reduced processing time and solvent

consumption. After the extraction process, macroporous resin column chromatography was

used to concentrate and purify the DHQ. Three resins were selected from fifteen macroporous

resins for further investigation of their performance. Among these resins, AB-8 resin exhibited

relatively better adsorption capacities and desorption ratios for DHQ. The ethanol volume
fraction of the solutions for sample loading and desorption, and flow rates for loading and

desorption were optimized for the macroporous resin column chromatography.

Keywords: Larix gmelinii; dihydroquercetin; homogenization; cavitation hybrid rotation

extraction; macroporous resin

1. Introduction

Dihydroquercetin (DHQ), 2-(3,4-dihydroxyphenyl)-2,3-dihydro-3,5,7-trihydroxy-

4Hbenzopyran-4-one), also known as taxifolin and vitamin P, is an important natural flavone

[1].

The pharmacologic activities of DHQ, including its inhibition or activation of enzymatic activity, and

antioxidant [2,3], anti-radiation [4], antiviral [5], and anti-tumor activities [6], are mainly attributed to its

phenolic groups. DHQ is a stronger antioxidant than common synthetic or natural antioxidants. DHQ is

safe for use by pregnant women and does not induce malformation, mutation, or allergies in the fetus

[7].

Larix gmelinii is a deciduous conifer of the Pinaceae family in the Larix genus [8]. It is native

to cooler temperate regions in the Northern Hemisphere, particularly to the boreal forests of

Russia and Canada [9]. In Northeast China, L. gmelinii comprises a large proportion of the

trees found in forests [10]. Because of its unique wood physical features of L. gmelinii, such

as its stiffness, straight grain, and resistance to corrosion, L. gmelinii has been applied

extensively in papermaking, building, and furniture manufacturing [11]. Consequently, large

quantities of waste, such as logging slash, and bucking and processing residues, are

generated every year. L. gmelinii is rich in DHQ [12], which could be of use in finding

applications for this waste.

 DHQ is usually extracted using traditional methods, such as Soxhlet extraction (SE), heat

reflux extraction (HRE), and ultrasonic or accelerated solvent extraction, which are generally

time consuming and have high energy consumption [13]. Supercritical fluid extraction has

been applied to DHQ, but has high equipment costs [14]. Enzymatic water extraction is an

efficient method for extraction of DHQ [12], but the high cost of enzyme treatment limits

extensive application of this method on industrial scales. Therefore, a cheap, simple, and

convenient extraction method is required for the extraction of DHQ. Negative pressure

cavitation extraction is a simple, eco-friendly, and scalable technology, which has been

successfully used for the extraction of natural products [15-21]. In this process, cavitation can

trigger the concentration of fluid energy and the drastic oscillation of the onflow, which

expands the cells, disrupts the cell walls, and produces high mass transfer velocities between

the cell wall and intracellular compounds. Consequently, the target natural products are

transferred from the cell to the extraction solution. The vacuum cavitation extraction method

has high extraction efficiency.

Homogenization is a conventional physical milling method, which facilitates the release of

components from a material into a solvent with high-velocity mechanical shearing, stirring,

fluid cutting action, and crushing without application of heat or pressure. This method has

been successfully used for the extraction of alkaloids, isoflavones, and pigments [22-24].

However, to the best of our knowledge, homogenization has not been combined cavitation

extraction with homogenization-assisted hybrid rotation extraction (HCHRE) to extract DHQ

from plant tissues.

 This aim of this study was to develop an improved HCHRE to extract DHQ from the xylem

of L. gmelinii. This involved pretreating the plant materials by homogenization and then

subjecting the homogenate to cavitation hybrid rotation extraction. Several extraction

parameters, such as the ethanol volume fraction, homogenization time, the liquid–solid ratio

for homogenization, cavitation extraction time, and vacuum strength were optimized to

increase the extraction yield. High-performance liquid chromatography (HPLC) analysis was

applied for the determination of DHQ during the extraction procedure. After extraction, the

crude extracts were purified by adsorption on a macroporous resin, and various macroporous

resins were tested for this step. Furthermore, the mechanism of homogenization treatment

and cavitation hybrid rotation extraction was explained from the angle of the formation,

growth, and collapse of the bubbles.

2. Experimental

2.1. Reagents and materials

2.1.1. Materials

L. gmelinii wood chips were sourced from the Greater Khingan Mountains (Heilongjiang

Province, China) and authenticated by Jian Li academician from the College of Material

Science and Engineering, Northeast Forestry University (Harbin, China). The wood chips

were dried in the shade with ventilation, and the moisture content was 12%.

2.1.2. Chemicals

DHQ reference standard (purity 98%) was purchased from the National Institute for the

Control of Pharmaceutical and Biological Products (Beijing, China). Deionized water used in
all of the experiments was prepared using a Milli-Q system (Millipore, Billerica, MA). HPLC

grade acetonitrile and acetic acid of were purchased from J&K Chemical Ltd. (Shanghai,

China). All other chemicals were of analytical grade and purchased from Beijing Chemical

Reagents Co. (Beijing, China). All solutions were filtered through 0.45-μm membranes

(GuangFu Chemical Reagents Co., Tianjin, China) before use.

2.1.3. Quantitative analysis of DHQ by HPLC

The HPLC system was equipped with a 1525 binary pump, a 717 automatic column

temperature control box, a 2487 ultraviolet-visible detector, and a 717 automatic sample

handling system (Waters, Milford, MA). Chromatographic separation was conducted on a HiQ

Sil C18 reversed-phase column (4.6 mm  250 mm, 5 µm, KYA TECH Corp., Tokyo, Japan).

An isocratic elution was performed using a mixture (18:82, v/v) of acetonitrile and an aqueous

solution of 1.0% acetic acid as the mobile phase. The mobile phase flow rate, injection

volume, and column temperature were 1.0 mL/min, 10 µL, and 25 °C, respectively. DHQ was

detected at 294 nm. Its retention time was 24 min, and the total run time was 30 min. The

equation for the calibration curve for the determination of DHQ was Y = 3.1755  107 X +

2.5993  104 (r = 0.9999). The linear range for DHQ was 0.03–0.50 mg/mL, and the liquid

chromatography of DHQ standard and the larch wood sample was shown in Figure 1.

Figure 1

2.2. Apparatus

We constructed a laboratory-scale apparatus or HCHRE using a 200 W homgenizer

(HANUO-JJ2, Shanghai Hannuo Instrument Co. Ltd., Shanghai, China), a glass cylinder
(internal volume, 1.5 L; diameter, 4.0 cm), a flow meter (Sierra 50L, Shanghai Siya Detection

Instrument Co. Ltd., Shanghai, China ), and a water circulating vacuum pump (SHB-Ш, Great

Wall Scientific Industry and Trade Co. Ltd., Zhengzhou, China). Pretreated mixtures of L.

gmelinii and the extraction solvent were added to the extraction column. Air was introduced

through a pump from the bottom of the extraction column, and the flow rate was measured

using the flow meter.

2.3. HCHRE procedure

The two main steps in the extraction procedure were pretreatment of the solvent and plant

material in the homogenizer, and extraction of DHQ from the homogenate in the negative

pressure cavitation hybrid rotation extraction apparatus. L. gmelinii wood chips (1.00 g) were

extracted with a 10 mL/g liquid–solid ratio and 10 min homogenization pretreatment process.

This was followed by cavitation hybrid rotation extraction for 30 min with a liquid–solid ratio of

9 mL/g. The ethanol volume fraction, liquid–solid ratio for homogenization, homogenization

time, liquid–solid ratio for cavitation extraction, and the pressure were optimized for the

HCHRE. After each extraction, the extraction solution (2.0 mL) was filtered through a 0.45-μm

nylon membrane and then analyzed by HPLC.

2.4. Comparison with traditional methods

HCHRE was compared with the traditional extraction methods of SE, HRE, and room

temperature soaking extraction (RTSE). The extraction time, extraction temperature, and

liquid–solid ratio for the extractions were as follows: 24 h, 95 °C, and 10 mL/g for SE; 6 h, 95

°C, and 10 mL/g for HRE; and 72 h, 20 °C, and 30 mL/g for RTSE.

2.5. Macroporous resin enrichment of DHQ

2.5.1. Screening of macroporous resins

Fifteen macroporous resins (Bochum Tanjin Resin Technology Co., Ltd.) were screened

using static adsorption tests. Before use, 3.0 g of each resin was hydrated resin. Then, each

hydrated resin was mixed with 150.0 mL of the crude extract in a 250-mL Erlenmeyer flask.

The flasks were then shaken (100 rpm) for 12 h at 25 °C to reach adsorption equilibrium.

Then, the supernatants were analyzed by HPLC to calculate the adsorption capacity of each

resin. Next, the resins were washed thoroughly with deionized water and desorption was

performed using aqueous ethanol (10.0 mL, 95% v/v). The flasks were shaken at 180

rpm/min for 6 h at 25 °C. The supernatants after desorption were also analyzed by HPLC.

Each experiment was performed three times.

The absorption capacities and desorption ratios were used to select resins and plot kinetic

curves for the adsorption and desorption. The absorption capacities (Qe, milligrams adsorbed

per gram of resin [mg/g resin]) and desorption ratios (Qd, %) were calculated using the

following equations:

Qe = (C0 − Ce) V0/m, (1)

where C0 and Ce are the initial solute concentration (mg/mL) and that at equilibrium,

respectively; V0 is the initial volume (mL) of crude extract; and m is the dry mass (g) of the

resin.

Qd = Cd Vd /m 100 (2)

where Cd, Vd, and m are the solute concentration for desorption (mg/mL), the desorption

solution volume (mL), and the dry mass of the resin (g), respectively.
2.5.2. Dynamic adsorption and desorption

Dynamic adsorption and desorption tests were conducted in a glass column (height–

diameter ratio, 15:1; column diameter, 20 mm) filled with the selected resin. A sample solution

with an appropriate concentration was loaded onto the column at a suitable flow rate and the

DHQ concentration was determined by HPLC. After reaching the adsorption saturation point,

the column was washed thoroughly with deionized water and then eluted with an aqueous

solution of ethanol. The DHQ concentration in desorption solution was determined by HPLC.

3. Results and discussion

3.1. Single factor experiments

3.1.1. Effect of the volume fraction of ethanol

Ethanol is frequently used for extraction because of its good penetration capacity to raw

materials, low boiling temperature, low cost, and safety. The extraction yield of DHQ is

affected by the volume fraction of ethanol and ethanol solutions with a low volume fraction will

probably not be sufficient for DHQ dissolution. As the volume fraction of ethanol was

increased from 20% to 60%, the DHQ extraction yield increased obviously (Figure 2a). After

this, a gradual increase in the extraction yield was found when the volume fraction of ethanol

was increased above 60%. The higher volume fraction of ethano increased molecular affinity

between solvent and solute, and the diffusion of the solvent into the solute, which can be

effectively improved by the use of homogenate. However, the extraction yield of DHQ is

declined slightly because of the “similar dissolve mutually theory”. Therefore, an ethanol

volume fraction of 60% was used in subsequent experiments.

3.1.2. Effect of the liquid–solid ratio in homogenization

 The liquid–solid ratio plays a crucial role in the extraction procedure. High liquid–solid ratios

could result in a tedious extraction process and needless solvent consumption, whereas low

liquid–solid ratios could lead to the incomplete dissolution of the solute in the solvent. In the

present study, several liquid–solid ratio ratios were evaluated for the homogenization. The

extraction yield of DHQ increased gradually as the liquid–solid ratio was increased from 8 to

10 mL/g (Figure 2b). This could be because a higher liquid–solid ratio allowed for better

contact between the raw materials than a lower liquid–solid ratio. Further increases in the

liquid–solid ratio resulted in only slight changes in the extraction yields. Therefore, 10 mL/g

was selected as the optimum liquid–solid ratio for subsequent experiments.

3.1.3. Effect of the homogenization time

To investigate the effect of the homogenization time, experiments were performed with

homogenization for 4, 6, 8, 10, and 12 min (Figure 2c). The extraction yield of DHQ was very

low at the first 4 min of homogenization, which indicates that a certain length of time is

required for homogenization to sufficiently decrease the particle size of the plant material and

increase its contact with the extraction solvent. When the homogenization time was increased

from 4 to 8 min, the extraction yield of DHQ increased dramatically, but further increases in

the homogenization time did not produce obvious improvements in the extraction yields. The

subsequent filtration step also increased in difficulty as the particle size became very small.

Therefore, a homogenization time of 8 min was selected for subsequent experiments.

3.1.4. Effect of the liquid–solid ratio for cavitation extraction

The influence of the liquid–solid ratio for cavitation extraction was investigated using ratios

of 6, 7, 8, 9, and 10 mL/g. The extraction yield of DHQ slowly increased as the liquid–solid
ratio was increased from 6 to 9 mL/g (Figure 2d). When the liquid–solid ratio was increased

above 9 mL/g, the DHQ extraction yield remained constant. A high liquid–solid ratio could

increase the complexity of the extraction process and solvent consumption. Therefore, a

liquid–solid ratio of 9 mL/g was selected for cavitation extraction.

3.1.5. Effect of the cavitation extraction time

To study the effect of the cavitation extraction time on the extraction yield of DHQ,

experiments were performed with extraction times of 5, 15, 25, 35, and 45 min. In these

experiments, the homogenization time was 8 min, the liquid–solid ratio for homogenization

was 10 mL/g, the volume fraction of ethanol was 60%, and the liquid–solid ratio for cavitation

extraction was 9 mL/g. When the cavitation extraction time was increased from 5 to 35 min,

the extraction yield of DHQ increased (Figure 2e). Further increases in the cavitation time

dramatically decreased the DHQ extraction yield. This occurred because a longer cavitation

time resulted in dissolution of more impurities, which greatly decreased the DHQ content in

the solution. Hence, 35 min was selected as the optimum time for cavitation extraction.

Figure 2

3.2. Comparison of HCHRE with traditional methods

The HCHRE method was compared with traditional methods (Table 1). Higher extraction

yields were obtained by HCHRE than the other methods. Although the extraction yield of DHQ

obtained with RTSE (DHQ content in extracts was 4.09±0.02 mg/g) was only slightly lower

than that obtained with HCHRE, HCHRE took only 0.72 h, including 8 min for homogenization

pretreatment and 35 min for cavitation hybrid rotation extraction, for extraction of DHQ (DHQ

content in extracts was 4.50% ± 0.02 mg/g). By contrast, although the energy consumption
of RTSE was the lowest, it took 72 h, and also had higher solvent consumption. HCHRE had

higher extraction yields than SE (DHQ content in extracts was 2.80% ± 0.01 mg/g) and HRE

(DHQ content in extracts was 3.32% ± 0.02 mg/g), and also reduced solvent and energy

use compared with these methods. In summary, HCHRE is an efficient, low energy (0.2

kW·h), and fast method that utilizes the synergetic effects of homogenization and cavitation

hybrid rotation.

Table 1

3.3. The homogenization mechanism

Mechanical effects: includes the generation of turbulence, liquid circulation currents and

shear stresses. Homogenization has strong mechanical forces and liquid shear forces that

result in good pulverization efficiency, due to fluid flow and impingement of high velocity liquid

on solid wall. Additionally, homogenization thoroughly mixes the matrix with the solvent. It can

also promote solvent penetration into the inner matrix. This leads to transfer of the target

compounds from the matrix to the solution [25]. In this process, the plant material particle size

decreased as the homogenization time increased (Figure 3), which is beneficial for mass

transfer because small particles have high specific surface areas exposed to the extraction

solvent [26-28] and shorter path lengths for mass transfer [29]. However, the longer

homogenization time was also leading the impurities dissolve. Thus, the homogenization time

should be less than 10 min. Meanwhile, turbulence effect in the solvent created in the

homogenizer decreases the thickness of the liquid membrane on the solid particle surface. It

also results in an increase in DHQ dissolved.

Figure 3
3.4. The cavitation hybrid rotation mechanism

After homogenization, the cell walls of the plant material were broken and the specific

surface area increased, which accelerated diffusion and mass transfer of DHQ from L.

gmelinii for cavitation hybrid rotation extraction. Vapor pressure and surface tension are the

two key factors that impact the cavitation intensity at a specific distance from the transducers

and generally cavitation intensity decreases as vapour pressure and surface tension

increases [30]. In this, the cavitation effect is generated by negative pressure. In cavitation

hybrid rotation extraction there are solid, liquid, and gas phases that interact in the following

four areas of the system (Figure 4): bubble formation, axis air flow, mixing and rotation, and

turbulence. In bubble formation, small bubbles are generated because of negative pressure,

and these bubbles continuously rise and form a highly unstable gas–liquid–solid system. In

the axis air flow area of the system, the bubbles collide with the liquid and solid phases and

form small droplets or small droplets containing solid particles. Contact between liquid

droplets occurs frequently, and this is promoted by increased flow of the bubbles, which

increases the probability of collisions and the mass transfer rate between the phases. This

step also establishes concentration gradients in the liquid phase with dissolution of DHQ. The

majority of the mass transfer occurs in the mixing and rotation area of the system. Here, the

bubbles expand rapidly, solid particles are separated quickly from the liquid phase with little

impact from efflux, and the bubbles, which in pressure gradient area expand and collapse

instantaneously, cause strong efflux. Due to limited “space” for the bubbles to expand, most of

the bubbles collapse asymmetrically on the glass wall of the equipment, resulting in significant

liquid circulation currents coupled with intense turbulence [31]. This makes the solid and liquid
phases combine and separate, the bubbles rapidly collapse to minimum volume, and the

dissolution of DHQ from solid phases to liquid phases reached balance.

Microstreaming resulting from stable cavitation has been shown to produce stresses sufficient to disrupt

cell membranes [32]. The mechanism proposed is the onset of turbulence, which creates vortices in

whose proximity the shear rates are higher than those throughout the bulk of the liquid [33]. In the

turbulence area of the system, the liquid membrane around the bubbles exchanges material quickly,

and the rising bubbles rapidly expand and then disintegrate near and at the liquid level because of a

sharp reduction in pressure. The turbulence effects indicated that there is large growth in the size of the

bubble/cavity, which also results in significant magnitudes of pressure pulse at the time of the collapse

[34]. Micro jets generated by the disintegration of the liquid phase membrane around the bubbles causes

disruption of the bubbles, vacuoles, and droplets, which cause results in disintegration of the bubbles

and mass transfer [15]. However, the longer cavitation time was also leading the impurities dissolve.

Thus, the cavitation time should be less than 40 min.

Figure 4

3.5. Static adsorption and desorption on macroporous resins for the enrichment of DHQ

3.5.1. Screening of macroporous resins

The adsorption capacities and desorption ratios of 15 macroporous resins were calculated

to select suitable resins for the enrichment and purification of DHQ from crude extracts of L.

gmelinii. The adsorption capacities of DHQ on AB-8, HPD-100A, and D-101 resins were

obviously higher than those on the other resins, and the desorption ratios of DHQ from AB-8,

HPD-200L, and D-101 resins were considerably higher than those for the other resins (Table

2). These results are in line with the properties of the resins (e.g. resin polarities, average
pore diameters, and surface areas) and the chemical characteristics of the compound of

interest. The AB-8, HPD-200L, and D-101 resins were selected for further study of the

adsorption of DHQ.

Table 2

3.5.2. Kinetics curves for static adsorption

Static adsorption of DHQ on the selected resins was evaluated at 25 °C. Kinetics curves for

the static adsorption of DHQ were plotted (Figure 5a), and showed that the AB-8 and D101

resins had high adsorption rates for DHQ in the first 3 h and then reached equilibrium. The

HPD-200L resin reached equilibrium after 4 h. Therefore, the AB-8 and D101 resins provide

more efficient adsorption of DHQ than the HPD-200L resin.

3.5.3. Kinetics curves for static desorption

Desorption of DHQ on the resins selected in Section 3.5.1 was also evaluated (Figure 5b).

The order of the resins from highest to lowest desorption ratio was AB-8 > D-101 > HPD200L.

AB-8 and D-101 had relatively higher adsorption rates of DHQ and the shorter times (2

h) to reach equilibrium than HPD-200L.

Taking into consideration the saturation adsorption, adsorption capacities, and desorption

ratios, AB-8 resin was selected as the optimum resin for subsequent experiments.

Figure 5

3.5.4. Adsorption isotherms of DHQ

Adsorption isotherms of DHQ on AB-8 resin were investigated at four different

temperatures. At the same initial concentration, the adsorption capacities decreased as the
temperature was increased from 25 to 55 °C. The rate of adsorption was slower than the rate

of desorption, which implied that the adsorption process was an exothermic process.

Therefore, the adsorption process should be conducted at room temperature. The initial

concentrations of DHQ were 0.9296, 0.4647, 0.2323, 0.1161, and 0.0580 µg/mL (Figure 6).

The parameters for the Langmuir and Freundlich adsorption isotherms are shown in Table S1,

and the Langmuir and Freundlich equations were used for linear fitting and to describe how

the solute interacted with the resins.

Figure 6

Table S1

3.5.5. Langmuir adsorption isotherm of DHQ

The Langmuir isotherm is the most well-known and frequently used isotherm for the

adsorption of solutes from solution. The Langmuir isotherm can be expressed as follows:

Qe =Qm Ce/(K + Ce) (3)

where Qe is the adsorption capacity, Ce (µg/mL) is the concentration of solute in solution

(liquid phase) at equilibrium, K is the Langmuir constant, and Qm is the empirical constant.

The equations and the correlation coefficients of the Langmuir adsorption isotherm at

different temperatures are summarized in Table S2. The results indicated the adsorption

process was a monomolecular layer adsorption. It also implied that the adsorption process

was an exothermic process. From these results, 25 °C was selected as the optimum

adsorption temperature.

3.5.6. Freundlich adsorption isotherm of DHQ

 The Freundlich adsorption isotherm is an empirical equation that is used for physical and

chemical adsorption with non-ideal systems. The Freundlich adsorption isotherm can be

expressed as follows:

Qe = K Ce1/n, (4)

or by the following equation:

lnQe = lnk + 1/n lnCe (5)

where K is the Freundlich constant, which is an indicator of adsorption capacity; and 1/n is

an empirical constant related to the magnitude of the adsorption driving force.

The equations and the correlation coefficients of the Freundlich adsorption isotherm at

different temperatures are summarized in Table S2. The results indicated that the adsorption

process was a monomolecular layer adsorption. The Freundlich equation correlation

coefficient of DHQ at 25 °C (R2 = 0.9946) was higher than that of the Langmuir equation at 25

°C (R2 = 0.9703). Therefore, the Freundlich adsorption isotherm describes the adsorption

behavior of DHQ on AB-8 resin better than the Langmuir isotherm.

Table S2

3.6. Effects of the ethanol volume fraction on sample loading and desorption

3.6.1. Effect of the ethanol volume fraction on absorption

To determine the optimum ethanol volume fraction for loading the sample, the same mass

of DHQ was dissolved in ethanol solutions with different volume fractions (10%, 20%, 30%,

40%, and 50%). As the ethanol volume fraction increased, the adsorption of DHQ decreased (Figure

7a). This probably occurred because of competitive adsorption between ethanol and DHQ, which have
similar polarities. A solvent with higher polarity will increase the absorption more than one with lower

polarity. Notably, the adsorption capacity with an ethanol volume fraction of 10% was lower than that

obtained with an ethanol volume fraction of 20%. This could be because the dissolution of DHQ

decreased as the solvent polarity increased, and this decreased the DHQ content of the sample solution.

Therefore, an ethanol volume fraction of 20% was selected for loading the sample.

3.6.2. Effect of the ethanol volume fraction on the desorption

To select a suitable desorption solution, ethanol solutions with different volume fractions

(50%, 60%, 70%, 80%, and 90%, v/v) were investigated. As the ethanol volume fraction

increased from 50%–80%, the desorption rate of DHQ increased (Figure 7b). When the

ethanol volume fraction was increased above 80%, only a slight improvement in the

desorption rate was observed. Therefore, to minimize solvent consumption, an ethanol

volume fraction of 60% was selected as optimum for DHQ desorption.

Figure 7

3.7. Dynamic adsorption and desorption on the AB-8 resin column

3.7.1. The effect of the loading flow rate on dynamic adsorption

The dynamic adsorption of DHQ on the AB-8 resin was investigated at 25 °C. The leakage

curves at different loading flow rates (Figure 8a) showed that loss of DHQ increased as the

flow rate was increased from 3 to 5 bed volume (BV)/h. This probably occurred because low

loading flow rates allowed for diffusion of the sample solution. However, if the flow rate is too

low, the experimental time will be too long. Generally, the breakthrough point can be defined

as 10% of the ratio of the solute concentration in the eluate to the solute concentration in the
loading solution [35]. Therefore, the optimal flow rate for loading the sample solution was 4

BV/h.

3.7.2. Effect of the flow rate for elution in the desorption process

The elution flow rate in desorption is an important factor, and we investigated three flow

rates (1 BV/h, 2 BV/h and 3 BV/h) with 60% ethanol. Elution curves were constructed (Figure

8b) and showed that higher flow rates increased the elution DHQ. With flow rates of 1 BV/h, 2

BV/h, and 3 BV/h, the elution volumes were 5, 7, and 11 BV, respectively. However,

increased elution may result in insufficient desorption of DHQ. To reduce solvent

consumption, increase the desorption ratio, and reduce the production of solvent waste, a

flow rate of 2 BV/h was selected as the optimum flow rate. Fractions (4.0 mL) of the

desorption solution were collected, concentrated in a rotary evaporator, dried under vacuum,

and then analyzed by HPLC. Finally, DHQ crystals (purity 65%) were obtained.

Figure 8

4. Conclusions

In this study, a HCHRE method was developed for extraction of DHQ from L. gmelinii

wood. The crude extracts were purified by macroporous resin column chromatography for

DHQ enrichment. Single factor experiments were conducted to determine the optimum

conditions for DHQ extraction. HCHRE is more efficient, faster, and has lower energy

consumption than traditional methods. Under the optimum extraction conditions, the HCHRE

has high extraction efficiency for DHQ and the DHQ content in extracts is 4.50±0.02mg/g.

Furthermore, the mechanisms of homogenization and cavitation hybrid rotation extraction

were examined in detail. Fifteen macroporous resins were tested for concentration and
purification, and three resins (D101, HPD-200L, and AB-8) were selected for further

screening. Among these resins, AB-8 is the optimum resin based on its adsorption capacities

and desorption ratios for DHQ. The optimum operating parameters are as follows: ethanol

volume fraction for loading the sample solution, 20%; ethanol volume fraction for desorption,

60%; loading flow rate, 4 BV/h; and desorption flow rate, 2 BV/h. The adsorption behavior of

DHQ on the AB-8 resin is described better by the Freundlich adsorption isotherm than the

Langmuir isotherm.

Acknowledgments

This study was supported by the National Key Technology R&D Program (Grant No.

2015BAD14B06), the postdoctoral scientific research developmental fund of Heilongjiang

Province, 2016 (Grant No. LBH-Q16001), the Research Start-up Funding of Introduce Talents

in Northeast Forestry University (Grant No. YQ2015-02), the National Natural Science

Foundation of China (Grant Nos. 31500467, 31570567), and the Natural Science Foundation

of Heilongjiang Province for Young Scholar (Grant No. QC2015034).

References

[1] S.M. An, H.J. Kim, J.E. Kim, Y.C. Boo, Flavonoids, Taxifolin and Luteolin Attenuate

Cellular Melanogenesis despite Increasing Tyrosinase Protein Levels, Phytother. Res. 22

(2008) 1200-1207.

[2] P. Trouillas, C. Fagnère, R. Lazzaroni, C. Calliste, A. Marfak, J.L. Duroux, A theoretical

study of the conformational behavior and electronic structure of taxifolin correlated with

the free radical-scavenging activity, Food Chem. 88 (2004) 571-582.

[3] A. Maroziene, R. Kliukiene, J. Sarlquskas, Inhibition of phthalocyanine-sensilized

photohemolysis of human erythrocytes polyphenolic antioxidants: description of

quantitives structure-activity relationships, Cancer Lett. 157 (2000) 39-44.

[4] N. Sugihara, T. Arakawa, M. Ohnishi, Anti-and pro-oxidative effects of flavonoids on

metalinduced lipid hydroperoxide-dependent lipid peroxidation in cultured hepatocytes

loaded with alpha-linolenic acid, Free Radic. Biol. Med. 27 (1999) 1313-1323.

[5] S.C. Chu, Y.S. Hsieh, J.Y. Lin, Inhibitory effects of flavonoids on moloney marine

leukemia virus reverse transcriptase, J. Nat. Prod 55 (1992) 179-183.

[6] S. Kawaii, Y. Tomono, E. Katase, Effect of citrus flavonoids on HL-60 cell differentiation,

Anticancer Res. 19 (1999) 1261–1269.

[7] S. Zu, L. Yang, J. Huang, C. Ma, W. Wang, C. Zhao, Y. Zu, Micronization of taxifolin by

supercritical antisolvent process and evaluation radical scavenging activity, Int. J. Mol.

Sci. 13 (2012) 8869–8881.

[8] D.S. Gernandt, A. Liston, Internal transcribed spacer region evolution in Larix and

Pseudotsuga (Pinaceae), Am. J. Bot. 86 (1999) 711–723.

[9] M.C. Gros-Louis, J. Bousquet, L.E. Pâques, N. Isabel, Species-diagnostic markers in

Larix spp. based on RAPDs and nuclear, cpDNA, and mtDNA gene sequences, and their

phylogenetic implications, Tree Genet. Genomes 1 (2005) 50–63.

[10] J. Zhang, Z.Z. Li, Composition Study of Phenol Compounds of Chinese Larch Tree, J.

Cellulose Sci. and Technol. 9 (2001) 16–20.

[11] C. Ma, L. Yang, W. Li, J. Yue, J. Li, Y. Zu, Ultrasound-assisted extraction of

arabinogalactan and dihydroquercetin simultaneously from Larix gmelinii as a

pretreatment for pulping and papermaking, PloS one 9(2014), 1–21.

[12] Y. Wang, Y. Zu, J. Long, Y. Fu, S. Li, D. Zhang, J. Li, M. Wink, T. Efferth, Enzymatic

water extraction of taxifolin from wood sawdust of Larix gmelinii (Rupr.) Rupr and

evaluation of its antioxidant activity, Food Chem. 126 (2011) 1178–1185.

[13] S.P. Pietarinen, S.M. Willfoer, F.A. Vikstrom, B.R. Holmbom, Aspen knots, a rich source

of flavonoids, J. wood Chem. Technol. 26 (2006) 245-258.

[14] J. Peng, G. Fan, Y. Chai, Y. Wu, Efficient new method for extraction and isolation of three

flavonoids from Patrinia villosa Juss. by supercritical fluid extraction and high-speed

counter-current chromatography, J. Chromatogr. A 1102 (2006) 44-50.

[15] Y. Kong, Z.F. Wei, Y.J. Fu, C.B. Gu, C.J. Zhao, X.H. Ya, T. Efferth, Negative-pressure

cavitation extraction of cajaninstilbene acid and pinostrobin from pigeon pea [Cajanus

cajan (L.) Millsp.] leaves and evaluation of antioxidant activity, Food Chem. 128 (2011)

596–605.

[16] M.M. Yan, C.Y. Chen, B.S. Zhao, Y.G. Zu, Y.J. Fu, W. Liu, T. Efferth, Enhanced

extraction of astragalosides from Radix Astragali by negative pressure

cavitationaccelerated enzyme pretreatment, Bioresour. Technol. 101 (2010) 7462–7471.

[17] D.Y. Zhang, Y.G. Zu, Y.J. Fu, M. Luo, W. Wang, C.B. Gu, C.J. Zhao, J. Jiao, T. Efferth,

Enzyme pretreatment and negative pressure cavitation extraction of genistein and

apigenin from the roots of pigeon pea [Cajanus cajan (L.) Millsp.] and the evaluation of

antioxidant activity, Ind. Crops Prod. 37 (2012) 311–320.

[18] M.H. Duan, W.J. Xu, X.H. Yao, D.Y. Zhang, Y.H. Zhang, Y.J. Fu, Y.G. Zu,

Homogenateassisted negative pressure cavitation extraction of active compounds from

Pyrola incarnata Fisch. and the extraction kinetics study, Innov. Food Sci. Emerg. 27

(2015) 86–

93.

[19] D.Y. Zhang, Y.G. Zu, Y.J. Fu, M. Luo, C.B. Gu, W. Wang, X.H. Yao, Negative pressure

cavitation extraction and antioxidant activity of biochanin A and genistein from the leaves

of Dalbergia odorifera T. Chen, Sep. Purif. Technol. 83 (2011) 91–99.

[20] D.Y. Zhang, X.H. Yao, M. Luo, C.j. Zhao, Y.J. Fu, Optimization of negative pressure

cavitation–microwave assisted extraction of yellow horn seed oil and its application on

the biodiesel production, Fuel 166 (2016) 67–72.

[21] M. Luo, L.Q. Yang, X.H. Yao, F.S. Mu, D.Y. Zhang, Z.Y. Song, Q. Qiao, Y.J. Fu, Y.G. Zu,

Optimization of enzyme-assisted negative pressure cavitation extraction of five main

indole alkaloids from Catharanthus roseus leaves and its pilot-scale application, Sep.

Purif. Technol. 125 (2014) 66–73.

[22] W.G. Shi, Y.G. Zu, C.J. Zhao, Y. Lei, Homogenate extraction technology of

camptothecine and hydroxycamptothecin from Camptotheca acuminata leaves, J. For

Res. 20 (2009) 168–170.

[23] X.Y. Zhu, H.M. Lin, J. Xie, S.S. Chen, P. Wang, Homogenate extraction of isoflavones

from soybean meal by orthogonal design, J. Sci. Ind. Res. 70 (2011) 455–460.
[24] X.Y. Zhu, Y.l. Mang, F.q. Shen, J. Xie, W.k. Su, Homogenate extraction of gardenia

yellow pigment from Gardenia Jasminoides Ellis fruit using response surface

methodology, J. Food Sci. Technol. 51 (2014) 1575–1581.

[25] M.H. Duan, W.J. Xu, X.H. Yao, D.Y. Zhang, Y.H. Zhang, Y.J. Fu, Y.G. Zu,

Homogenateassisted negative pressure cavitation extraction of active compounds from

Pyrola incarnata Fisch. and the extraction kinetics study, Innov. Food Sci. Emerg. 27

(2015) 86-

93.

[26] A. Brachet, P. Christen, J.L. Veuthey, Focused microwave-assisted extraction of cocaine

and benzoylecgonine from coca leaves, Phytochem. Anal. 13 (2002) 162–169.

[27] C.S. Eskilsson, E. Bjöklund, Analytical-scale microwave-assisted extraction, J.

Chromatogr. A 902 (2000) 227-250.

[28] A.H. El-Ghorab, A.F. Mansour, K.F. El-massry, Effect of extraction methods on the

chemical composition and antioxidant activity of Egyptian marjoram (Majorana hortensis

Moench), Flavour Frag. J. 19 (2004) 54-61.

[29] R.M. Banik, D.K. Pandey, Optimizing conditions for oleanolic acid extraction from

Lantana camara roots using response surface methodology, Ind. Crop. Prod. 27 (2008)

241-248.

[30] H. Li, L. Pordesimo, J. Weiss, High intensity ultrasound-assisted extraction of oil from

soybeans, Food Res. Int. 37 (2004) 731–738.

[31] S.R. Shirsatha, S.H. Sonawanea, P.R. Gogate, Intensification of extraction of natural

products using ultrasonic irradiations—A review of current status, Chem. Eng. Process.

53 (2012) 10–23.

[32] G. Scherba, R.M.Weigel, W.D. O’Brien, Quantitative assessment of the germicidal

efficacy of ultrasonic energy, App. Environ. Microb. 57 (1991) 2079.

[33] P. R. Gogate, A. M. Kabadi, A review of applications of cavitation in biochemical

engineering/biotechnology, Biochem. Eng. J. 44 (2009) 60–72.

[34] P. R. Gogate, Hydrodynamic Cavitation for Food and Water Processing, Food

Bioprocess Technol. 4 (2011) 996–1011.

[35] W. Liu, S. Zhang, Y.G. Zu, Y.J. Fu, W. Ma, D.Y. Zhang, Y. Kong, X.J. Li, Preliminary

enrichment and separation of genistein and apigenin from extracts of pigeon pea roots by

macroporous resins, Bioresour. Technol. 101 (2002) 4667-4675.

Figure captions

Figure 1 Liquid chromatography of DHQ standard and the larch wood sample

Figure 2 Effect of volume fraction ethanol (a), homogenate liquid-solid ratio (b), homogenate

time (c), and cavitation liquid-solid ratio (d), cavitation time (e) on the extraction efficiency

of DHQ

Figure 3 Schematic diagram of homogenate extraction mechanism

Figure 4 Schematic diagram of cavitation hybrid rotation extraction mechanism

Figure 5 Adsorption kinetic curves (a) and desorption kinetic curves (b) for DHQ

Figure 6 Langmuir (a) and Freundlich (b) adsorption isotherm

Figure 7 Effect of sample solvent on DHQ adsorption amount (a) and elute solvent on DHQ

desorption amount (b)

Figure 8 Leakage curves (a) and dynamic desorption curves (b) with different current velocity
Table list

Table 1 Comparison of the yields and purities of DHQ extracted with different methods

Extraction time Extraction Solvent Energy consumption DHQ content in

Method temperature (°C) volume (mL) (kW·h) extracts
(h)
SE 24 95 300 0.5×24=12 2.80
HRE 6 95 300 0.5×6=3 3.32
RTSE 72 20 900 0 4.09
HCHRE* 0.72* 20 540 0.75×0.13+0.18×0.58=0.2 4.50
homogenate
* This extraction time includes 8 min (0.13 h) for homogenization and 35 min (0.58 h) for
cavitation hybrid rotation extraction.
Table 2 Physical properties and absorption characteristics of macroporous resins

Specific surface Average Adsorbing Desorption

Name Type
area (m2/g) pore size capacity (mg/g) ratios (%)

HPD-100 Nonpolar 650-700 8.5-9.0 11.4 70.8

HPD-100A Nonpolar 650-700 9.5-10.0 16.8 67.4

HPD-300 Nonpolar 800-870 5.0-5.5 9.0 79.9

HPD-700 Nonpolar 650-700 8.5-9.0 9.4 65.5

HPD-5000 Nonpolar 550-600 10.0-11.0 4.9 74.6

D101 Weak-polar 400-600 10.0-12.0 16.2 81.9

AB-8 Weak-polar 480-520 13.0-14.0 17.4 92.5

HPD-400 Middle polar 500-550 7.5-8.0 10.8 80.6

HPD-200L Middle polar 500-550 8.0-9.0 15.8 87.2

HPD-400A Middle polar 500-550 8.5-9.0 12.7 72.8
HPD-450 Middle polar 500-550 9.0-11.0 13.6 67.9

HPD-750 Middle polar 650-700 8.5-9.0 10.7 63.4

HPD-500 Strong polar 500-550 5.5-7.5 6.5 62.5

HPD-600 Strong polar 550-600 8.0 6.3 62.9

HPD-850 Strong polar 1100-1300 8.5-9.5 6.8 67.4

Figr-1
Figr-2
Figr-3
Figr-4
Figr-5 Figr-6
Figr-7
Figr-8