Effects of water and methanol on synthesis of polyoxymethylene dimethyl ethers from dimethoxymethane and paraformaldehyde

https://doi.org/10.1016/j.ces.2021.117057Get rights and content

Highlights

  • Effects of water and methanol were studied in detail for synthesis of PODEn.

  • Method was proposed to calculate CH2O conversion in presence of water/methanol.

  • A complete SF model was proposed that was applicable to different reactants.

  • Water and methanol greatly decreased PODE3-5 yield and should be removed.

Abstract

Water and methanol are two unavoidable by-products for synthesis of polyoxymethylene dimethyl ethers (PODEn) from dimethoxymethane (DMM) and paraformaldehyde (PF), but their effects have not been clearly studied. As DMM decomposes with water and is formed with methanol, the decomposition ratio of DMM is a key quantity to describe the effects of water and methanol. Using this quantity, a simple and accurate method was proposed for calculating CH2O conversion and a complete model of Schulz-Flory (SF) distribution was established in the presence of water/methanol, both of which were validated experimentally. Then the effects of water and methanol were investigated in detail. The presence of water and methanol had complicated effects on equilibrium of the reaction, and both significantly decreased the yield of PODE3-5. These results showed that both water and methanol should be removed from the recycling stream in an industrial plant to enhance the efficiency of the reactor.

Graphical abstract

Comparison of simplified and complete models in terms of the calculated values of ae (Hollow symbols: simplified model; Solid symbols: complete model).

  1. Download : Download high-res image (82KB)
  2. Download : Download full-size image

Introduction

As the exhaustion of petroleum is becoming one of the major concerns in the 21st century (Gowdy and Juliá, 2007), oxygenated fuels, which can be produced from natural gas (Aasberg-Petersen et al., 2011), coal (Höök and Aleklett, 2010), and biomass (Li et al., 2016), provide an alternative to partly substitute petroleum fuel. In addition, oxygenated fuels can promote the combustion efficiency of fuel (Li et al., 2017), which can reduce the exhaust gas emission.

Polyoxymethylene dimethyl ethers (PODEn, CH3O(CH2O)nCH3, where n ≥ 1) are a kind of efficient oxygenated fuel used as diesel additives. For simplicity and consistency, PODE1 was used to refer to dimethoxymethane (DMM) in this work. Compared to other diesel additives, such as methanol (MeOH) (Wang et al., 2013), dimethyl ether (Zhao et al., 2005), and DMM (Zhu et al., 2009), PODEn are better due to their higher cetane number, higher flash and boiling points and more similar physical properties to diesel (Baranowski et al., 2017, Boyd, 1961). Among all PODEn products, PODE3-5 are ideal diesel additives because their properties are very close to that of diesel (Liu et al., 2017).

PODEn can be produced in different approaches (Hackbarth et al., 2018), such as oxidative coupling of dimethyl ether (Gao et al., 2018, Zhang et al., 2016) and polymerization of formaldehyde (Wang et al., 2015, Zhang et al., 2014). The latter is efficient to produce PODEn in industry, using terminal group supplier, such as methanol and DMM, and methoxyl group supplier, such as paraformaldehyde (PF), formaldehyde solution and trioxane, to react over acid catalysts. Because water had a negative effect on this reaction, DMM and PF are preferred reactants for industrial production (Zheng et al., 2013).

In the PODEn synthesis process, water and methanol cannot be absolutely avoided due to water content in solid PF, which will react with DMM to produce methanol. If not strictly separated, water and methanol will be recycled to the reactor from the separation unit. Water and methanol have significant influences on the CH2O conversion, PODEn product distribution, and PODE3-5 yield.

Firstly, the presence of water and methanol make it difficult to calculate the CH2O conversion. In the literature, the CH2O conversion was usually replaced by trioxane conversion (Fu et al., 2015, Li et al., 2015a, Li et al., 2015b, Wang et al., 2014, Wu et al., 2015a, Wu et al., 2015b, Wu et al., 2014, Xue et al., 2017) or PF conversion (Liu et al., 2017, Liu et al., 2018b, Zheng et al., 2016, Zheng et al., 2013, Zheng et al., 2015a). However, the CH2O and PF/trioxane conversions are not exactly the same because some CH2O molecules are released from PF/trioxane and dissolved as formaldehyde in solution without further reaction. With the addition of water or methanol, the amount of dissolved CH2O molecules increases significantly, resulting in a large calculation error if the CH2O conversion is simplified as PF/trioxane conversion. Meanwhile, the presence of water/methanol caused decomposition/formation of DMM, respectively, which released/consumed CH2O molecules, as shown in reaction (R1). This reaction was usually ignored in the literature when calculating the CH2O conversion.DMM + H2O ↔ CH2O + 2CH3OH

Secondly, the presence of water and methanol affected the product distribution of PODEn. Although it was proved that the equilibrium product distribution followed the Schulz–Flory (SF) distribution when using DMM and PF as reactant (Zhao et al., 2013, Zheng et al., 2015b), it is unclear whether the SF distribution model is still correct with the addition of water and methanol. According to reaction (R1), the addition of water or methanol will change the mole balance equations of the system, which are used to derive the SF distribution model (Zheng et al., 2015b). Therefore, the SF model should be restudied at the presence of water and methanol.

Thirdly, the addition of water and methanol affected the yield of PODE3-5. It was reported that water and methanol affected the dissolution of PF and the extent of the formation of PODEn, but it is unclear how they affect the equilibrium state.

Although there were good dynamic and thermodynamic models in the literature (Hahnenstein et al., 1995, Schmitz et al., 2015), these models were proposed for the homogeneous system and could not by directly used to the heterogeneous system using PF and DMM as reactants. This work aimed to investigate the effects of water and methanol on synthesis of PODEn from PF and DMM. An accurate method was proposed for calculating the CH2O conversion and the SF distribution model was refined considering the effects of water and methanol. Then the effects of water and methanol on PODE3-5 yield were investigated in detail. Based on these results, suggestion was provided for the industrial production process of PODEn in terms of the control of water/methanol concentration.

Section snippets

Materials

DMM (analytic reagent grade, AR), and PF (analytic reagent grade, AR) were purchased from Alfa Aesar-Johnson Matthey. Methanol (analytic reagent grade, AR) was purchased from Shanghai Titan Scientific Co., Ltd. The NKC-9 cation exchange resins were dry resins of H+ type provided by Tianjin Bohong Resin Technology Co., Ltd. The resins were pre-treated by a drying oven at 80 °C for 24 h and stored in a desiccator. The water content of PF was ∼3 wt%, and the mass fraction of PF used in every

Verification of conversion calculation

With the addition of water or methanol, the calculation of CH2O conversion became much more complicated. Experiments were carried out to verify the calculation method proposed in Section 2. In these experiments, an enough amount of water or methanol was fed to the reaction system to ensure all the solid reactant was dissolved after reaction, so that the titration results represented the overall amount of unreacted CH2O molecules after reaction.

The feeding ratio and product analysis results of

Conclusions

Water and methanol were two by-products in the PODEn synthesis process, which would be recycled to the reactor if they are not strictly removed in the separation units. In this work, the influences of water and methanol on the synthesis of PODEn from DMM and PF was investigated both theoretically and experimentally, leading to the following conclusions:

  • (1)

    Water and methanol caused decomposition and formation of DMM, respectively, and the decomposition ratio of DMM was a key quantity to describe

CRediT authorship contribution statement

Fang Liu: Conceptualization, Methodology, Validation, Investigation, Formal analysis, Resources, Data curation, Writing – original draft, Visualization. Ran Wei: Investigation. Tiefeng Wang: Conceptualization, Supervision, Project administration, Funding acquisition, Writing – review & editing.

Declaration of Competing Interest

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

Acknowledgement

The authors thank the financial supports by the National Key Research and Development Program of China (No. 2018YFB0604804).

References (36)

Cited by (0)

View full text