Pilot Scale Electrolysis of Peroxodicarbonate as an Oxidizer for Lignin Valorization

A pilot scale plant at Technology Readiness Level (TRL) 6 comprising an electrochemical ex-cell continuous production of sodium peroxodicarbonate and a thermal depolymerization plug flow reactor for kraft lignin conversion is established. Due to the labile nature of the “green” oxidizer peroxodicarbonate, special attention must be paid to the production parameters in order to optimize its use. A simplified design model describing steady-state and transient operations is formulated and finally validated against experimental data from the electrolysis setup. Design trade-offs are visualized, and their impact on specific energy consumption is evaluated. The pilot plant was operated for a 20-month period for more than 1200 h on-stream. Optimized process conditions result in vanillin yields of 8 wt % and thus prove the successful scale-up.

1 List of symbols, superscripts and abbreviations

Setups
The pilot plant (Figure S1) was installed at the multi-phase flow laboratory at SINTEF in Trondheim.It consists of a feed system for raw materials, an electrochemical reactor setup for continuous production of peroxodicarbonate, and a thermal depolymerisation reactor for oxidation and thermal depolymerisation of lignin.The reactor system is designed for continuous operation, while reactor product separation and upgrading is performed in a semi-batch separation process.

Tiller Multiphase Laboratory
Drawing title:

TEU
Designed by: Checked by: Approved by: 102019323 Project No: 1/1   The electrochemical cell is powered by a power supply manufactured by plating electronic GmbH.The power station pe4606 provides up to 300 A at up to 10 V with a maximum net effect of 2800 W. Cooling downstream of the electrochemical cell was provided by a fusion-bonded plate heat exchanger, type 14-30H, manufactured by Alfa Laval Nordic AS.
The cooling capacity was 3 kW with a minimum temperature approach of 3 o C. The cooling water temperature was kept above 8 o C to avoid the precipitation of sodium carbonate.The circulation pump was a model MD-30RZM-2020N from Iwaki Co., Ltd. with a magnetic drive.The maximum capacity was 15 L/min.Noteworthily, that all data were recorded over more than a year, without significant change in the performance of the cell.Key design parameters are displayed in Table S4.The thermal depolymerization reactor (Figure S3) was built based on the following concept: Figure S3: Thermal depolymerisation reactor design as a plug flow reactor with internal static mixers for radial mixing.Feed locations are at the entrance of reactor section 1 and 3.

Process control system
A process control system (Implemented by VisonTech AS in IGSS from Schneider Electronics) has been established for the lignin feed and reactor systems in the pilot which enable 24 hours unmanned operation of the unit.The control system is linked to a Labview based datalogger service which stores all process data to a SQL database which is part of the lab infrastructure.
The human machine interface (HMI) for the peroxodicarbonate and the thermal depolymeri-sation reactor are shown in Figure S4 and S5.
Control parameters for peroxodicarbonate reactor (electrolyzer) are: • Start/stop signal to circulation pump PX64807 • Start/stop of the power supply unit (PSU) • Set-point for output current from the PSU -given as current density • Set-point for temperature control of the circulation loop via TIC64301 • Start/stop of the peroxodicarbonate feed pump PX64806 • Set-point for volumetric feed flow rate from PX64806 [ml/min] Figure S4: Process control HMI for peroxodicarbonate reactor system.
Input parameters are: • Distribution of peroxodicarbonate feed to either reactor section 1 or 3 via FC64913 • Percent of time period where peroxodicarbonate goes towards reactor section 3 (default is zero) • Length of time period (default 20 min) • Heat input to heating zone 1, 0-1900 W • Heat input to heating zone 2, 0-1900 W

Materials
The Kraft lignin used was the Lineo ® Classic product from Stora Enso.It is a technical lignin derived from spruce and pine and was isolated by means of the LignoBoost process at the Sunila mill, Kotka, Finland.
Sodium carbonate for the electrochemical production of peroxodicarbonate was industrial grade calcinated and granulated soda purchased from Solberg Industri AS.See Table S5 chemical composition and physical data.The soda was delivered in 25 kg bags and mixed with tap water in a 1000 l IBC container with stirring.def2-TZVPP (def2/JK) for the anionic species have been used and B3LYP was selected as hybrid functional.4][5][6][7][8][9][10][11][12][13][14] For all calculations, the conductor-like polarizable continuum model (CPCM) was used with water as solvent, which is implemented in ORCA.The anionic geometries and atom coordinates have been setup based on common bond angles and bond length.The geometry was then optimized using iterative single-point energy calculations.

ELF calculations
Based on the results of the DFT calculation, the ELF was calculated.The ELF offers a method to localize paired electrons in chemical systems and hence, it strongly correlates with the chemical intuitive concept of bonds and lone pairs.The pair probability distribution (P ) to find an electron at a given spacial position (⃗ r) with the same spin as the probe at a given distance (s) will be zero if s = 0 due to the Pauli principle: [15][16][17][18][19][20][21][22][23][24][25][26][27] If P is expanded to a power series and under the assumption that s is small, a positional dependent parameter C(⃗ r) is analyzed.Within the Hartree-Fock theory and the consideration of the Salater determinant as the simplest form of a wave function, C(⃗ r) can be described using the occupation number (n i ), the orbital (Φ) and the electron spin density (ρ).
For practical reasons, Becke and Edgecombe 17 introduced the homogeneous electron gas (h) as a reference system on the one hand and a scaling of the function to a range between [0..1] to be able to compare results from different calculations: The ELF value converges to 1 at regions where the probability to find paired electrons is very high (e.g.paired electrons, lone pairs) and will be 0.5 in a homogeneous electron gas.Usually, the ELF will be analyzed in a topological manner, where local maxima of the functions are called attractors and spacial regions are created by gradient calculations in the ELF where such a region is called ELF-basin (Ω).The number of electrons ( N ) populating such a spacial region is calculated by integrating the electron density in the ELF-Basins.
ELF calculations, topological analysis and electron density integration were performed with Multiwfn.

Figure S1 :
Figure S1: Process flow diagram of the electrochemical lignin depolymerization plant.

Figure S2 :
Figure S2: P&ID of the electrochemical reactor configuration with internal recirculation loop and feed and bleed reservoir.

Table S1 :
List of symbols.

Table S2 :
List of subscripts.

Table S3 :
List of abbreviations.

Table S4 :
Key design parameters for the pilot plant depicted.

Table S5 :
Chemical and physical data for sodium carbonate.Aqueous sodium hydroxide for the dissolved lignin feed stream was purchased from Solberg Industri AS at 46 wt% concentration and was stored in an isolated (at 30 o C) IBC container, prior to mixing with water and lignin in the 200 l feed tank.See TableS6for chemical and physical data.

Table S6 :
Chemical and physical data for aqueous sodium hydroxide.
The DFT calculations have been performed with the ORCA program package 1 in version 4.2.1 (x86-64, Intel Core i7-1255U/12 cores).The def2 basis sets, especially def2-TZVP and Oxidant concentration was determined by iodometric titration where the oxidants H 2 O 2 and Na 2 C 2 O 6 oxidize iodide to iodine in the presence of sulphuric acid and with molybdate as catalyst.Na 2 C 2 O 6 + 2 H 2 SO 4 + 2 KI I 2 + K 2 SO 4 + Na 2 SO 4 + 2 H 2 O + 2 CO 2

Table S8 :
Base case operating conditions, system dimensions, and model parameters.Solving the least square minimization gives the following model parameters, which all are found to significantly describe the variance in the model data: .0010.03 0.03 0.90 1.00 and trace elements in an urban road setting in Trondheim, Norway: Re-visiting the chemical markers of traffic pollution.Science of The Total Environment 2019, 649, 703-711, doi: 10.1016/j.scitotenv.2018.08.299.