Bioresource Technology Reports 11 (2020) 100471

Contents lists available at ScienceDirect

Bioresource Technology Reports

journal homepage: www.journals.elsevier.com/bioresource-technology-reports

An integrated process for combined microbial VOC reduction and effluent T

valorization in the wood processing industry
⁎,1
Martin Lindemanna,c, , Bernhard Widhalma,1, Thomas Kuncingerb, Ewald Srebotnikc
a
Kompetenzzentrum Holz GmbH – Wood K plus, Altenberger Straße 69, A-4040 Linz, Austria
b
Fritz Egger GmbH & Co. OG, Tiroler Straße 16, A-3105 Unterradlberg, Austria
c
Institute of Chemical, Environmental and Bioscience Engineering, Technische Universität Wien, Getreidemarkt 9, A-1060 Vienna, Austria

A R T I C LE I N FO A B S T R A C T

Keywords: This study focused on the utilization of medium-density fiberboard (MDF) effluent as a culture medium for
Volatile organic compounds Pseudomonas putida for reduction of volatile organic compounds (VOC) in pinewood. P. putida was successfully
MDF effluent valorization cultivated in the residual carbohydrate-rich effluent after purification by centrifugation, 30 kDa membrane
Pseudomonas putida filtration and recovery of bioactive compounds via adsorption onto polyvinyl polypyrrolidone (PVPP). Besides
Terpene biodegradation
monosaccharides, P. putida can metabolize glycerol, acetate, succinate, and citrate present in MDF effluent
Bioactive compounds
Lignocellulosic biomass
without addition of other nutrients. P. putida cultures applied to pinewood reduced emissions of α-pinene and
Δ3-carene by 80% and 50%, respectively, and β-pinene and α-terpinolene were reduced by 100% within 4 days.
The valorization of industrial process effluents for VOC reduction could assist the wood processing industry in
sustainably lowering VOC emissions of their products.

1. Introduction chemical/biological oxygen demand (COD/BOD) and eliminate toxic

substances at minimum cost. A key aspect of improving overall sus-
Wood processing industries must handle large amounts of process tainability for the industry is direct reuse of the effluent (Ashrafi et al.,
effluent, especially effluent from pretreatment of woody biomass. 2015). Current literature now focuses more on utilizing effluent
During the production of medium-density fiberboards (MDF), a large streams, for instance, by fermentation. Specific microorganisms can use
effluent stream is generated by squeezing water out of steam-pretreated multiple carbon sources to produce a desired product, while simulta-
softwood chips. This water is rich in total organic carbon (TOC) and neously reducing COD/BOD. For example, acidogenic fermentation of
contains various carbohydrates, polyphenols, and organic acids. The pulp mill effluents can be the initial step for production of poly-
latter include colloidal fatty and resin acids stabilized by hemicelluloses hydroxyalkanoates (PHA) by producing volatile fatty acids (VFA)
and various cations (Otero et al., 2000). Effluent treatment is complex (Bengtsson et al., 2008). Queirós et al. (2014) showed that VFA pro-
and costly, requiring multiple steps such as sedimentation, flocculation, duced by acidogenic fermentation from hardwood spent sulfite liquor is
biological treatment, membrane filtration and reverse osmosis. Utili- a substrate for PHA production by a mixed microbial culture under
zation of currently unused process effluent stream would be beneficial aerobic conditions. Kaushik and Jadhav (2017) published an interesting
via reducing carbon load and thus decreasing oxygen demand during approach, utilizing pulping industry wastewater to produce electricity
biological treatment. In our previous work (Lindemann et al., 2020), we in a microbial fuel cell, with Pseudomonas fluorescens being the main
presented a method for recovery of valuable polyphenols (lignans and energy-producing bacterium.
stilbenes) from MDF effluent by adsorption onto a commercial poly- A vast number of Pseudomonas species are ubiquitous in soil and
vinyl polypyrrolidone (PVPP) resin. However, after adsorptive removal water and can adapt to challenging environments, including adverse
of polyphenols, a residual complex matrix of carbohydrates, small or- conditions, such as high and low temperatures and poor nutrient
ganic acids, and salts is left that would still need treatment before availability. Further, pseudomonads are known for their ability to
disposal or reuse. metabolize a wide range of substrates, such as hydrocarbons, aromatic
Traditionally, the main goal when handling wastewaters from compounds and terpenes (Mirpuri et al., 1997, Poblete-Castro et al.,
pulping and biomass processing industries was solely to decrease 2012, Yoo et al., 2001). For example, Bicas et al. (2008) described the

⁎
Corresponding author at: Kompetenzzentrum Holz GmbH – Wood K plus, Altenberger Straße 69, A-4040 Linz, Austria.
E-mail address: martin.lindemann@tuwien.ac.at (M. Lindemann).
1
Both authors contributed equally to this work.

https://doi.org/10.1016/j.biteb.2020.100471
Received 28 April 2020; Received in revised form 2 June 2020; Accepted 4 June 2020
Available online 06 June 2020
2589-014X/ © 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

biotransformation of monoterpenes in two Pseudomonas species, and stopped before polyphenol breakthrough. Polyphenols were then re-
Widhalm et al. (2016) showed that an adapted mixed culture of Pseu- covered by elution with methanol. Full details on preparation and the
domonas metabolized aldehydes and terpenes in pinewood. Greatly development of the adsorption system can be found in Lindemann et al.
reduced emissions of these compounds were observed after only 3 days. (2020).
α-Pinene, the major terpene in pinewood (Kleinheinz et al., 1999), is
classified as an irritant and hazardous to the environment according to 2.2. Chemicals
the Globally Harmonized System of Classification, Labeling and
Packaging of Chemicals (GHS/LPC). Large amounts of wood products Chemicals for medium preparation, monoterpenes and toluene D8
made from pinewood in buildings may thus negatively affect indoor air standard and sugar/sugar alcohol standards, (D-(+)-glucose, D-
quality due to elevated concentrations of aldehydes and terpenes. (+)-xylose, D-(+)-arabinose, D-(+)-galactose, D-(+)-mannose, D-
Chronic exposure to these VOC may be a cause of “sick-building-syn- (−)-fructose, D-(+)-pinitol), trimethylchlorosilane (TMCS, ≥99.0%),
drome” (Makowski and Ohlmeyer, 2006; Wilke et al., 2013). To avoid a heneicosanoic acid (> 99%) and betulin (> 98%) were obtained from
negative impact on indoor air quality and consequently on human Sigma Aldrich (St. Louis, MO, USA). Anion multi-element standards
health, various strategies to lower total VOC emission levels have been (Certipur® Anionen-Multielement-Standardlösung I (F−, PO43−, Br−)
pursued. Manninen et al. (2002) and McGraw et al. (1999) found that and II (Cl−, NO3−, SO42−)) and cation multi-element standard
terpene emissions from pinewood are less after heat treatment than (Certipur® Kationen-Multielement-Standardlösung VI) (NH4+, K+,
after air-drying. Chemical degradation of terpenes from pinewood with Na+, Ca2+, Mg2+) were purchased from Merck KGaA (Darmstadt,
sodium carbonate, sodium sulfite, sodium hydroxide and mixtures of Germany). Organic acid standards (acetate, propionate, formate, suc-
these substances has also been investigated by Roffael et al. (2015). cinate, and citrate) were purchased from Carl Roth GmbH & Co.KG
Furthermore, biodegradation and biotransformation with different (Karlsruhe, Germany). PVPP-RS (Divergan® RS, cross-linked poly-1-(2-
fungi in liquid culture (Lee et al., 2015; Qi et al., 2002) and reduction of oxo-1pyrrolidinyl) ethylene for use in regenerable processes, average
VOC from waste gases with biofilters containing fungi (Van Groenestijn particle size 80–100 μm) was kindly provided by BASF SE
and Liu, 2002) has been reported. Recently, Widhalm et al. (2018) (Ludwigshafen, Germany). Milli-Q-Water was prepared according to
showed that pretreatment of pine wood strands with a mixed culture of DIN ISO 3696 (1991). All other chemicals used were of analytical
Pseudomonas and the ascomycete Penicillium nigricans resulted in a grade.
significant decrease of α-pinene, β-pinene and Δ3-carene emissions
from oriented strand boards (OSB). 2.3. Microorganism, medium preparation, and growth conditions
In this study, we demonstrate the suitability of MDF process effluent
originating from steam pretreatment of wood chips as a medium to Pseudomonas putida NCIMB 10684 was obtained from the National
cultivate P. putida, after recovering resin, fatty acids, and polyphenols. Collection of Industrial, Food and Marine Bacteria (Aberdeen, Scotland,
Cultures of P. putida can reduce aldehydes and terpenes in pinewood UK) and maintained in Petri dishes on Standard Nutrient Agar No.1
strands. No additional carbon source or other nutrients are required. (Carl Roth GmbH, Karlsruhe, Germany). This strain was selected due to
Reusing MDF effluent for the cultivation of VOC reducing bacteria its superior ability to degrade α-pinene at a relatively low temperature
would allow the manufacture of wood products with reduced emissions optimum as revealed by a search of public databases (Widhalm et al.,
from treated pine wood strands sustainably and at low cost. 2016).
P. putida was grown at an optimal temperature of 25 °C in 50 mL of
2. Material and methods sterilized M9 minimal salts medium in sealed 250 mL Erlenmeyer flasks
on a rotary shaker at 200 rpm for 72 h (Widhalm et al., 2018) to obtain
2.1. Origin of MDF process effluents and general preparation procedures a liquid phase culture for further tests. One liter of M9 contained 6 g
Na2HPO4, 3 g KH2PO4, 0.5 g NaCl, and 1 g NH4Cl. Each flask was also
Process effluents were obtained from a MDF plant in Wismar, supplemented with 0.1 mL of 1 M aqueous MgSO4, 0.125 mL trace
Germany (Egger Holzwerkstoffe Wismar GmbH & Co. KG) between elements solution, and 0.25 mL of a 20% (w/v) glucose solution. This
April and September 2017 (PE-A: April 2017; PE-B: May 2017; PE-C: medium was used for several subsequent tests. For the trace element
August 2017; PE-D: September 2017). TOC values of the sampled ef- solution, 2.7 g FeCl3, 0.2 g ZnCl2, 0.2 g CoCl2, 0.2 g Na2MoO4, 0.1 g
fluents varied between 1 and 2 g/L (data not shown). Briefly, the MDF CaCl2, 0.13 g CuCl2, 0,05 g H3BO3 and 10 mL (37%) HCl were diluted
process involves steaming of wood chips at about 120 °C at elevated with sterile water to 100 mL. For all experiments, 1 mL portions of this
pressure, followed by squeezing water out of the steamed wood chips pre-culture were inoculated into additional flasks for cultivation in
using a stuffing screw. The material is then fed into a defibrator where various media.
it is mechanically broken down into fibers, while the water is dis-
charged as process effluent exhibiting TOC values between 1 and 2 g/L 2.4. Growth experiments of P. putida in liquid culture
(data not shown). Each effluent sample was sampled into a sterile
container at around 80 °C directly behind the stuffing screw. The con- The presence of inhibiting substances and bioavailable carbon
tainer was immediately sealed to avoid contamination. Spruce and pine sources other than monosaccharides in MDF process effluents were
are the primary woods used for MDF production at the Wismar plant, examined with P. putida liquid cultures in original effluent after re-
but small percentages of fir and beech are also used. The amount of moval of fibers and colloids and before adsorption of polyphenols, and
softwood comprises > 90% of the wood mix, although a precise com- in the flow-through of the polyphenol adsorption column without the
position cannot be assigned because only rough daily averages are addition of other nutrients. Original effluent or column flow-through
available. were sterilized before inoculation but pH was not adjusted and varied
The work-up procedure for the process effluent involved several from 5.6 to 7.0. Subsequently, myo-inositol, citrate, acetate, succinate,
steps. Briefly, fibers and non-colloidal particles were removed from glycerol, and xylitol were tested as individual carbon sources at final
MDF process effluent by centrifugation. Tangential flow membrane concentrations of 0.01 M in M9 medium. For each carbon source, six
filtration using a 30-kDa cut-off membrane cassette was then performed 100 mL sterilized Erlenmeyer flasks containing 40 mL medium were
to remove colloidal particles, resins, and fatty acids. Adsorption and prepared, followed by the addition of 0.1 mL MgSO4 (1 M) and 0.125 μl
removal of polyphenols was performed using a medium pressure liquid trace elements solution. Three of the six flasks were further supple-
chromatography (MPLC) column (25 mm diameter) filled with PVPP- mented with 250 μl of a 20% (w/v) glucose solution. Each flask was
RS (5.8 g). The flow-through of the column was collected and loading inoculated with 2 mL of M9-grown P. putida pre-culture, capped with a

2
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

cotton plug and incubated at 200 rpm and 25 °C. 2.7. Analysis of carbohydrates in process effluent with HPAEC/PAD
Bacterial growth was determined by measuring optical densities of
bacterial cultures at a wavelength of 600 nm (OD 600) on a Shimadzu Sugars and sugar alcohols in different process effluent samples were
UV-1800 spectrophotometer (Shimadzu Corp., Kyoto, Japan) (Cheng determined by high-performance anion-exchange chromatography with
et al., 2013). When effluents exhibited variable background absorption pulsed amperometric detection (HPAEC-PAD) on a Dionex ICS 5000+
at 600 nm, the OD 600 value immediately after inoculation was sub- system (Thermo Scientific) with Dionex CarboPac PA1 guard and ana-
tracted from OD 600 values at later time points. Data thus represent lytical (2 × 250 mm) columns (Thermo Scientific) at 0.26 mL/min as
newly formed biomass. follows. Preconditioning: 200 mM NaOH/65 mM sodium acetate for
10 min, followed by water for 7 min. Separation: Isocratic elution in
water with post-column addition of 67 mM NaOH for 18 min. Injection
2.5. Pinewood treatment with P. putida and VOC sampling volume: 10 μl. Sugars were quantified by calibration of peak areas with
authentic standards (D-arabinose, D-galactose, D-glucose, D-mannose,
Pinewood strands with approx. dimensions 2.5 cm × 12 cm length D-xylose, D-fructose) run along with samples in each sequence.
and 0.5 cm thickness were obtained from Fritz Egger GmbH (Wismar, Sugar alcohols were not adequately separated with this method. A
Germany). Moisture content was 100% ± 10% (wwater/wdry wood) and sum parameter was therefore defined by quantifying sugar alcohols as
was determined for 10 random samples with a Sartorius MA 150 pinitol (the predominant sugar alcohol identified in process effluents)
moisture analyzer (Sartorius AG, Göttingen, Germany). equivalents of total peak area eluting between 1.5 min and 2.2 min.
For pinewood treatment with P. putida, pre-cultures grown for 72 h This fraction represents mainly sugar alcohols as confirmed by GC–MS
in M9 medium or various effluents were first conditioned for terpene as an orthogonal method. Column eluates from 1.5 min to 2.2 min of 5
degradation by adding 0.05 mL α-pinene and further incubated for consecutive injections were pooled, freeze-dried, derivatized, and
24 h. Conditioned bacterial cultures were then used to inoculate cru- analyzed by GC–MS according to Lindemann et al. (2020). The sugar
shed fresh pinewood strands. alcohols were identified by mass spectra using Wiley and NIST mass
Emission tests of pinewood strands under laboratory conditions spectral libraries and quantified as percentages of total peak area.
were initially performed in headspace vials and later on also in μCTE™
micro-chambers/thermal extractors (Markes International Ltd., 2.8. Analysis of anions, cations, and organic acids in process effluent with
Llantrisant, UK). Specifically, 2 g of crushed pinewood strands were IC
inoculated with 2 mL conditioned P. putida pre-culture and incubated in
20 mL headspace vials (Thermo Scientific, Vienna, Austria) at Inorganic ions and organic acids in process effluent samples were
25 °C ± 2 °C (Widhalm et al., 2018). Unsterile strands moistened with determined by ion chromatography (IC) on a Dionex ICS 5000+ system
2 mL of tap water served as controls. For a detailed evaluation of vo- (Thermo Scientific) equipped with conductivity detectors and an au-
latile substances, further experiments were performed in μCTE™ micro- tosampler (Dionex AS-DV) for simultaneous injection onto both anion
chambers. A micro-chamber consists of six parallel cells with a volume and cation columns via two 25 μl sample loops.
of 48 cm3 each, arranged in a heating block. Two grams of crushed Inorganic anions and organic acids were separated on a Dionex
pinewood strands were weighed into Erlenmeyer flasks, inoculated with IonPac AS11-HC analytical column (4 × 250 mm) with a Dionex ATV-
2 mL of bacterial solution, and incubated for defined treatment times. HC trap pre-column (9 × 75 mm) at 1.5 mL/min as follows. Gradient
Thereafter, strands were heat-treated at 103 °C ± 2 °C overnight, elution: 1.5 mM NaOH from 0 to 5 min, 1.5 mM–5 mM NaOH from 5
cooled in a desiccator and then put into micro-chambers. After a con- to10 min, 5 mM–23 mM NaOH from 10 to 18 min, 23 mM–50 mM
ditioning phase of 1 h at 23 °C, VOC were sampled on Markes C1-AAXX- NaOH from 18 to 20 min, 50 mM from 20 to 25 min, 50 mM–1.5 mM
5003 Tenax TA tubes (Markes International Ltd., Llantrisant, UK). from 25 to 27 min. Suppressor: Dionex AERS 500 (4 mm); suppressor
Sampling time was 30 min at an airflow rate of 50 mL/min. A one μl current 19 mA from 0 to 5 min, 83 mA from 5 to 18 min, 186 mA from
toluene D8 standard was added to each tube before sampling. 18 to 27 min.
Cations were separated on a Dionex CS12A analytical column
(4 × 250 mm) with a CG12A guard column (4 × 50 mm) at 1 mL/min.
2.6. Analysis of VOC emissions with GC–MS Separation: Isocratic elution in 20 mM methanesulfonic acid for 28 min.
Suppressor: Dionex CERS 500 (4 mm); Suppressor current 59 mA.
Wood samples in headspace vials were analyzed for volatilized All ions were quantified by calibration of peak areas with authentic
compounds by solid-phase microextraction and subsequent gas chro- standards of chloride, nitrate, sulfate, phosphate; acetate, propionate,
matography–mass spectrometry (SPME/GC–MS) (Stratev et al., 2015). formate, succinate, citrate; sodium, ammonium, potassium, magne-
A solution of cyclodecane dissolved in methanol (1:100) served as an sium, calcium.
internal standard. Five μl of this solution was added through the septum
to each headspace vial immediately before measurement. VOC were 3. Results and discussion
adsorbed on a Supelco divinylbenzene/carboxene/poly-
dimethylsiloxane SPME fiber inserted through the septum into the vial 3.1. Process effluent composition
for 25 min at 50 °C. Fiber desorption was performed on an Agilent
7890A/5975C GC–MS system with a CTC Combi PAL autosampler and The chemical composition of the soluble fraction of four process
splitless injection of the analytes into an Agilent 19091S-433 column effluent samples PE-A, PE-B, PE-C, and PE-D, collected over 6 months
with dimensions 30 m × 0.25 mm × 0.25 μm. Temperature program and after removing fibers, resin acids, fatty acids, and colloidal particles
and other separation parameters were set according to Stratev et al. (Lindemann et al., 2020), was characterized by determining the pre-
(2015). sence of essential nutrients for the growth of P. putida, particularly
Tenax tubes were analyzed following DIN ISO 16000-6 (2010) using inorganic anions and cations, monosaccharides, sugars alcohols, and
a TD-100 Thermodesorption Unit (Markes), connected to an Agilent organic acids (Fig. 1). All samples contained glucose, fructose, man-
7890A/5975C GC–MS system. Analytes were separated on a HP-PONA nose, arabinose, and galactose in varying proportions. The most ob-
column (Agilent) with the dimensions 50 m × 0.2 mm × 0.5 μm. The vious differences among effluent samples were high concentrations of
following temperature program was used: 3 min at 35 °C, 35 °C → glucose, mannose and particularly fructose in PE-A and PE-B, compared
160 °C at 10 °C min−1, 160 °C → 260 °C at 20 °C min−1, hold time at to PE-C and PE-D. Data obtained from 26 effluent samples collected
260 °C was 10 min. regularly over 3 years indicated seasonal variations with elevated

3
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

Fig. 1. Inorganic ions (a), organic acids (b), and sugars/sugar alcohols (c) identified in PE-A, -B, -C and -D.

monosaccharide and sugar alcohol levels during winter and spring

(November–May), and reduced levels during summer and autumn
(June–October). Hou (1985) observed similar seasonal fluctuations of
sugars in spruce trunk wood. PE-A and PE-B were collected in April and
May, while samples PE-C and PE-D were collected in August and Sep-
tember. Thus, and thus the observation likely reflects seasonal varia-
tion.
As another possible carbon source for P. putida relatively large
amounts of sugar alcohols were found in all samples. Sugar alcohols are
shown in Fig. 1 as a sum parameter because they eluted as a group of
unresolved peaks by the routine HPAEC method employed (data not
shown). However, GC–MS analysis of the HPAEC eluate comprising
these peaks confirmed that 75.6% of the peaks present in the total ion
chromatogram were sugar alcohols, particularly D-pinitol (34.8%),
glycerol (19.7%), D-arabitol (8.1%), D-mannitol (5.1%), myo-inositol
(4.2%), isopinitol (1.7%), inositol (0.8%), D-threitol (0.8%), and 2-
deoxyribitol (0.4%).
Inorganic anions and cations that are required for microbial growth, Fig. 2. Growth curves of P. putida in shake flasks in two different process ef-
particularly sodium, ammonium, magnesium, calcium, and phosphate, fluent samples PE-A and PE-B with and without (w/o) polyphenols (pp).
were present in all process effluent samples. Concentrations were si-
milar in all samples, except for nitrate. Nitrate was only found in PE-B after 10 h of incubation compared to citrate, succinate, and the control
and, at lower concentrations, also PE-A, but not in PE-C and PE-D. (glucose). For acetate, its lower C/mol ratio may explain slower growth
Acetate and citrate were major organic acids found in all process ef- since no further increase in OD 600 occurred after 24 h. For glycerol,
fluent samples. PE-A and PE-B contained twice the amount of acetate as we observed an extended lag-phase and slower growth rate during the
compared to PE-C and PE-D. Strikingly high concentrations of citrate first 8 h that could be due to time needed for elaborate upregulation of
and particularly succinate were detected in PE-B. Succinate was absent glycerol catabolic genes and downregulation of other possible routes for
in PE-C and present only in minor concentrations in PE-A and PE-D. carbon consumption (Nikel et al., 2014). These results are in good
Data collectively suggest that utilization of process effluents as a nu- agreement with Hintermayer and Weuster-Botz (2017) who in-
trient source for P. putida may require effluent composition monitoring vestigated batch growth with organic acids and glycerol as single
to support proper adjustment of cultivation conditions to ensure con- carbon sources. No investigated substance was completely inhibitory
sistent results. for growth of P. putida, because a distinct increase in OD 600 occurred
in all media supplemented with glucose after 24 h. However, the pre-
3.2. Cultivation of P. putida in process effluents and synthetic media sence of several substances such as acetate and particularly xylitol ex-
tended the lag-phase and thus the onset of bacterial growth in liquid
Hourly measurements of OD600 were used as an indicator of bio- culture. In contrast, mixtures of glucose with succinate or citrate ap-
mass during growth of P. putida in two carbohydrate-rich process ef- peared to accelerate microbial growth in comparison to glucose as
fluents PE-A and PE-B (Fig. 2). Data demonstrate for both effluent single carbon source, which indicates simultaneous consumption of
samples that polyphenols strongly inhibited the growth of P. putida. both carbon sources. These results provide an initial basis for predicting
Removal of polyphenols by adsorption onto PVPP abolished inhibition the impact of carbon source variability and complexity in process ef-
resulting in growth rates similar to those obtained with synthetic fluents on the growth rate of P. putida and allow compensation for
media. Adsorption onto PVPP was employed for preparing all process nutrient deficiencies or growth inhibition by supplementing with sti-
effluents utilized for subsequent P. putida fermentation. mulating nutrients or extending fermentation time.
In a second step, we monitored the growth of P. putida with single
carbon sources, apart from monosaccharides, that naturally occur in 3.3. Effect of pre-cultivation on subsequent terpene degradation in pinewood
process effluents, particularly organic acids and sugar alcohols, either by P. putida
alone (Fig. 3a) or in combination with glucose (Fig. 3b).
P. putida metabolized acetate, citrate, succinate, and glycerol as sole P. putida was pre-cultivated in M9 medium under standard condi-
carbon sources. However, no growth was observed with myo-inositol tions as previously described (Widhalm et al., 2017), as well as in un-
and xylitol. Glycerol and acetate both exhibited lower OD 600 values treated (PE-B-crude) and purified (PE-B) process effluent where

4
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

SPME/GC–MS as reduced emissions of major terpenes (Fig. 4). As ex-

pected from the cultivation study described in Section 3.2, process ef-
fluent without prior removal of polyphenols (PE-B-crude) did not show
any reduction of terpenes compared to the uninoculated control. The
pre-cultivation of P. putida in purified effluent lacking polyphenols (PE-
B), however, was successful and resulted in a drastic reduction of ter-
pene emission comparable to that observed with M9 medium alone.
Specifically, α-pinene, Δ3-carene, p-cymene, and limonene emissions
were reduced by 80%, 50%, 80%, and 65%, respectively, while β-
pinene and α-terpinolene were completely absent after treatment bac-
teria were pre-cultivated in purified PE-B effluent. The incomplete
(80%) reduction of α-pinene emission was the only obvious difference
compared to standard pre-cultivation in synthetic M9 medium. Al-
though pre-cultivation in M9 medium and process effluent PE-B both
included a conditioning step with α-pinene before application onto
pinewood, the plurality of other available carbon sources in PE-B may
have hindered the full adaption of the terpene degrading enzyme
system of P. putida. Thus, simultaneous metabolism of terpenes and
residual carbon sources in PE-B may have retarded α-pinene degrada-
tion, whereas P. putida pre-cultivated in M9 was probably more efficient
due to the absence of other carbon sources during the adaption phase as
shown by Widhalm et al. (2017).

3.4. Fractionation of process effluents by adsorption onto PVPP

Adsorptive removal of polyphenols (mainly lignans and stilbenes)

has proved essential for successful utilization of process effluents as
growth media for P. putida. Therefore, the adsorption process was stu-
died in more detail using a fixed-bed column packed with PVPP. Highly
polar substances present in process effluents, particularly carbohy-
drates, sugar alcohols, salts, and organic acids were not retained by
PVPP and were thus fully recovered in the flow-through as previously
shown (Lindemann et al., 2020). However, monitoring the flow-
through at 280 nm during continuous loading with process effluent also
revealed a partial breakthrough of UV absorbing substances as shown in
Fig. 3. Growth curves of P. putida in shake flasks in M9 with various carbon Fig. 5a for the loading cycle of effluent batch PE-C. Although UV ab-
sources, with (a) and w/o (b) glucose as additional carbon source. (SD of 3 sorption increased over time, no polyphenols (i.e., lignans and pino-
replicates for each data point is provided in the supplementary materials.) sylvin) were detected in the flow-through as determined by GC–MS
(data not shown). Polyphenols strongly bound to PVPP and were fully
recovered at the end of the loading cycle by elution with a small volume
of methanol followed by a washing step with NaOH (Lindemann et al.,
2020). Thus, the UV increase during loading may be ascribed to highly
polar substances that would not or only weakly bind to PVPP and
cannot properly be identified with GC–MS (data not shown) due to their
irregular structures. Potential candidates are glycosylated polyphenols
or low molecular weight (MW) lignin carbohydrate complexes (LCC;
Tarasov et al., 2018). The latter assumption is supported by comparison
of UV spectra (Fig. 5b) and MW distributions (Fig. 5c) of original ef-
fluent, column eluate, methanol eluate, and NaOH wash. UV spectra of
crude effluent and methanol eluate both showed a distinct maximum at
280 nm consistent with authentic polyphenol (7-Hydroxymatairesinol)
standard, while distinct maxima were virtually absent. MW distribution
analysis (Fig. 5b) of the methanol eluate revealed two distinct peaks at
443 Da and 743 Da. The major peak co-eluted with the polyphenol (7-
hydroxymatairesinol) standard. In contrast, crude effluent, flow-
through fractions, and particularly NaOH wash showed a polydisperse
(PD > 1.4) MW distribution comprising a large share of higher MW
Fig. 4. VOC emissions of pinewood strands treated with P. putida pre-cultivated
material > 1.000 Da. Judged by its strong binding to PVPP, higher MW
in process effluent PE-B with and without (w/o) polyphenols (pp) compared to material in the NaOH wash may represent relatively non-polar lignin
M9 and an uninoculated control. oligomers rather than polar LCC. To determine the effect of these un-
defined UV absorbing substances and other unidentified substances
possibly present in the flow-through, P. putida was pre-cultivated in
polyphenols have been removed. Pre-cultures were applied onto pine-
consecutive fractions of the column flow-through, followed by appli-
wood strands to judge the effects of pre-cultivation on terpene de-
cation onto pinewood strands for comparison of VOC reduction.
gradation. After 4 days of incubation, degradation was detectable by

5
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

Fig. 6. TVOC emissions of pine wood strands treated with P. putida pre-culti-
vated in three consecutive flow-through fractions obtained during adsorption of
polyphenols onto PVPP. Two different process effluent samples, PE-C and PE-D,
are compared with an uninoculated control.

subsequent VOC degradation. Reduced performance of fraction 2 that

did not further deteriorate with fraction 3, indicates growth inhibition
by weakly binding substances breaking through the column. If phenolic
lignin oligomers or LCC are indeed involved in the observed inhibition,
their removal or inactivation might further enhance the performance of
process effluent for pre-cultivation of P. putida. One technical possibility
would be pretreatment of process effluent with lignin polymerizing
enzymes such as fungal laccase (Srebotnik and Hammel, 2000).

4. Conclusions

We successfully demonstrate the suitability of MDF process effluent

as a cultivation medium for P. putida after adsorptive removal of in-
hibitory polyphenols. Additional carbon sources and minerals are not
required. Apart from sugars, several other carbon sources in MDF ef-
fluents are metabolized as a single carbon source as well as in combi-
nation with glucose. Further, application of P. putida biomass obtained
from pre-cultivation in process effluent efficiently reduced VOC emis-
Fig. 5. UV absorption of the flow-through during a loading cycle of MDF pro- sions from pinewood. Results collectively contribute to the develop-
cess effluent PE-C on an MPLC column packed with PVPP (a). Absorbance ment of an environmentally benign biotechnological process for the
spectra (b) molecular weight distributions (c) of effluent and various effluent wood industry.
fractions.
CRediT authorship contribution statement
3.5. Effect of pre-cultivation in consecutive process effluent flow-through
fractions on VOC reduction Martin Lindemann:Conceptualization, Investigation,
Methodology, Data curation, Visualization, Writing - original
Changes in the chemical composition of the flow-through overtime draft.Bernhard Widhalm:Conceptualization, Investigation,
during adsorption of polyphenols onto PVPP may affect VOC reducing Methodology, Data curation, Visualization, Writing - original
potential of P. putida on pinewood. Therefore, the flow-through of two draft.Thomas Kuncinger:Resources, Conceptualization, Project
independently processed effluents, PE-C and PE-D, was fractionated and administration.Ewald Srebotnik:Conceptualization, Visualization,
the first three fractions were used for pre-cultivation of P. putida, fol- Writing - review & editing, Supervision.
lowed by application of cultures onto pinewood strands (Fig. 5 for PE-
C). After two days of incubation, VOC emissions from the strands were Declaration of competing interest
measured under controlled conditions in test chambers by collecting
released VOC onto sorbent tubes followed by GC–MS. The complete The authors declare that they have no known competing financial
GC–MS dataset of individual VOC showing extensive reduction for all interests or personal relationships that could have appeared to influ-
detected volatile aldehydes and terpenes, including the most re- ence the work reported in this paper.
calcitrant, Δ3-carene (Widhalm et al., 2018), is provided as supple-
mentary material. In Fig. 6, all these emissions from pinewood strands, Acknowledgments
treated with P. putida conditioned to single eluate fractions, are sum-
marized and represented as total VOC (TVOC). A distinct reduction in The authors thank the Austrian government and the federal gov-
TVOC was always achieved irrespective of fraction, but fraction 1 from ernments of Upper Austria, Lower Austria and Carinthia for funding the
both effluents provided the best conditions for pre-cultivation and Kompetenzzentrum Holz GmbH – Wood K plus, and TU Wien University

6
M. Lindemann, et al. Bioresource Technology Reports 11 (2020) 100471

Library for financial support through its Open Access Funding 533–538.
Programme. Manninen, A., Pasanen, P., Holopainen, J., 2002. Comparing the VOC emissions between
air-dried and heat-treated Scots pine wood. Atmos. Environ. 36, 1763–1768.
McGraw, G.W., Hemingway, R.W., Ingram, Leonard L., Canady, C.S., McGraw, W.B.,
Appendix A. Supplementary data 1999. Thermal degradation of terpenes: camphene, Δ3-carene, limonene, and α-ter-
pinene. Environ. Sci. Technol. 33, 4029–4033.
Mirpuri, R., Jones, W., Bryers, J.D., 1997. Toluene degradation kinetics for planktonic
Supplementary data to this article can be found online at https:// and biofilm-grown cells of Pseudomonas putida 54G. Biotechnol. Bioeng. 53,
doi.org/10.1016/j.biteb.2020.100471. 535–546.
Nikel, P.I., Kim, J., de Lorenzo, V., 2014. Metabolic and regulatory rearrangements un-
derlying glycerol metabolism in Pseudomonas putida KT 2440. Environ. Microbiol.
References 16, 239–254.
Otero, D., Sundberg, K., Blanco, A., Negro, C., Tijero, J., Holmbom, B., 2000. Effects of
Ashrafi, O., Yerushalmi, L., Haghighat, F., 2015. Wastewater treatment in the pulp-and- wood polysaccharides on pitch deposition. Nord. Pulp Pap. Res. J. 15, 607–613.
paper industry: a review of treatment processes and the associated greenhouse gas Poblete-Castro, I., Becker, J., Dohnt, K., dos Santos, V.M., Wittmann, C., 2012. Industrial
emission. J. Environ. Manag. 158, 146–157. biotechnology of Pseudomonas putida and related species. Appl. Microbiol.
Bengtsson, S., Werker, A., Christensson, M., Welander, T., 2008. Production of poly- Biotechnol. 93, 2279–2290.
hydroxyalkanoates by activated sludge treating a paper mill wastewater. Bioresour. Qi, B., Moe, W., Kinney, K., 2002. Biodegradation of volatile organic compounds by five
Technol. 99, 509–516. fungal species. Appl. Microbiol. Biotechnol. 58, 684–689.
Bicas, J., Fontanille, P., Pastore, G., Larroche, C., 2008. Characterization of monoterpene Queirós, D., Rossetti, S., Serafim, L.S., 2014. PHA production by mixed cultures: a way to
biotransformation in two pseudomonads. J. Appl. Microbiol. 105 (6), 1991–2001. valorize wastes from pulp industry. Bioresour. Technol. 157, 197–205.
Cheng, X.-Y., Tian, X.-L., Wang, Y.-S., Lin, R.-M., Mao, Z.-C., Chen, N., Xie, B.-Y., 2013. Roffael, E., Schneider, T., Dix, B., 2015. Effect of oxidising and reducing agents on the
Metagenomic analysis of the pinewood nematode microbiome reveals a symbiotic release of volatile organic compounds (VOCs) from strands made of Scots pine (Pinus
relationship critical for xenobiotics degradation. Sci. Rep. 3, 1869. sylvestris L.). Wood Sci. Technol. 49.
DIN ISO 3696. Water for Analytical Laboratory Use: Specification and Test Methods. DIN Srebotnik, E., Hammel, K.E., 2000. Degradation of nonphenolic lignin by the laccase/1-
Standards Committee Materials Testing, Berlin, Germany, 1991. hydroxybenzotriazole system. J. Biotechnol. 81, 179–188.
DIN ISO 16000-6: Bestimmung von VOC in der Innenraumluft und in Prüfkammern, Stratev, D., Günther, E., Steindl, J., Kuncinger, T., Srebotnik, E., Rieder-Gradinger, C.,
Probenahme auf Tenax TA, thermische desorption und gaschromatographie mit MS/ 2015. Industrial waste water for biotechnological reduction of aldehyde emissions
FID. DIN Standards Committee Materials Testing, Berlin, Germany, 2010. from wood products. Holzforschung 69, 463–469.
Hintermayer, S.B., Weuster-Botz, D., 2017. Experimental validation of in silico estimated Tarasov, D., Leitch, M., Fatehi, P., 2018. Lignin–carbohydrate complexes: properties,
biomass yields of Pseudomonas putida KT2440. Biotechnol. J. 12, 1600720. applications, analyses, and methods of extraction: a review. Biotechnol. Biofuels 11,
Hou, W., 1985. Seasonal fluctuation of reserve materials in the trunkwood of Spruce 269.
[Picea abies (L.) Karst.]. J. Plant Physiol. 117, 355–362. Van Groenestijn, J.W., Liu, J.X., 2002. Removal of alpha-pinene from gases using bio-
Kaushik, A., Jadhav, S.K., 2017. Conversion of waste to electricity in a microbial fuel cell filters containing fungi. Atmos. Environ. 36, 5501–5508.
using newly identified bacteria: Pseudomonas fluorescens. Int. J. Environ. Sci. Widhalm, B., Ters, T., Srebotnik, E., Rieder-Gradinger, C., 2016. Reduction of aldehydes
Technol. 14, 1771–1780. and terpenes within pine wood by microbial activity. Holzforschung 70, 895–900.
Kleinheinz, G.T., Bagley, S.T., St. John, W.P., Rughani, J.R., McGinnis, G.D., 1999. Widhalm, B., Rieder-Gradinger, C., Kuncinger, T., Srebotnik, E., 2017. Biodegradation of
Characterization of alpha-pinene-degrading microorganisms and application to a terpenes for emission-reduced oriented strand boards (OSB). Holzforschung 71,
bench-scale biofiltration system for VOC degradation. Arch. Environ. Contam. 259–264.
Toxicol. 37, 151–157. Widhalm, B., Rieder-Gradinger, C., Kuncinger, T., Srebotnik, E., 2018. Biotechnological
Lee, S.-Y., Kim, S.-H., Hong, C.-Y., Kim, H.-Y., Ryu, S.-H., Choi, I.-G., 2015. approach for α-pinene, β-pinene, and Δ3-carene degradation in pine wood for re-
Biotransformation of (−)-α-pinene by whole cells of white rot fungi, Ceriporia sp. duced terpene emissions from Oriented Strand Boards. Int. Biodeterior. Biodegrad.
ZLY-2010 and Stereum hirsutum. Mycobiology 43, 297–302. 134, 103–109.
Lindemann, M., Rieder-Gradinger, C., Kuncinger, T., Srebotnik, E., 2020. Selective re- Wilke, O., Brozowski, F., Wiegner, K., Brauer, F., 2013. Measurement of VOC emissions
covery of polyphenols from MDF process waters by adsorption on a macroporous, from OSB boards and their assessment according to the AgBB scheme. UMID. 1, 5–11.
cross-linked pyrrolidone-based resin. Holzforschung 74 (2), 217–225. Yoo, S.K., Day, D., Cadwallader, K., 2001. Bioconversion of α- and β-pinene by
Makowski, M., Ohlmeyer, M., 2006. Influences of hot pressing temperature and surface Pseudomonas sp. strain PIN. Process Biochem. 36, 925–932.
structure on VOC emissions from OSB made of Scots pine. Holzforschung 60,