On the effect of wax content on paraffin wax deposition in a batch oscillatory baffled tube apparatus

https://doi.org/10.1016/j.cej.2007.04.018Get rights and content

Abstract

Deposition of paraffin wax is one of the major problems facing in the petroleum industries with the main implication being wax blockage in oil and gas pipelines especially for offshore production. The objectives of this work are to investigate the effect of applying oscillatory motion on wax deposition in an oscillatory baffled tube apparatus, a relatively new mixing technology that offers more uniform mixing and solid suspension than traditional devices; and are to understand the mechanism and kinetics of the wax crystallisation in accordance to the Avrami theory. The wax deposition was determined gravimetrically. The results indicate that the oscillatory motion has two opposite effects on the percentage of wax deposition: at low concentration of wax in solution, the presence of oscillation significantly reduces the wax deposition, e.g. 40–60% without the presence of any solvent or wax inhibitor; and completely prevents 100% wax gelation from occurring – the beneficial effect; at higher wax contents, however, the introduction of oscillatory motion not only promotes wax deposition, but also accelerates the crystal growth to achieve 100% wax deposition – the detrimental effect.

Introduction

Extraction of crude oil from offshore reservoirs usually endures serious problems such as the obstruction of pipelines. Oil enters the pipeline at 60–70 °C and because the ocean water is much colder (4 °C in deep waters); the oil is cooled causing heavy hydrocarbons to precipitate out as the oil flows through the pipeline. The precipitated wax deposits on the tube inner wall, forming a solid layer that narrows the flow passage and eventually reduces the flow rates [1]. Traditional methods of management, prevention and remediation have been established over many years which involve mechanical means such as pigging, thermal treatment such as circulation of warm liquid, and chemical treatment such as addition of solvent and/or dispersant [2].

Much of the research conducted in the area of wax deposition has been centred around predicting and modelling of wax deposition, see for examples, the prediction of flocculation from petroleum fluids [3]; the prediction of the viscosity of waxy oils [4], [5]; the improved thermodynamic model for wax precipitation [6]; the prediction of the phase behaviour of petroleum fluids [7]; the prediction of wax deposition for flow assurance [8]; and the new thermodynamic model validated with wax disappearance temperature [44].

Laboratory investigations on wax deposition have been carried out such as using a cold finger set-up [9], [10]; using flow loops set-up [11]; in multiphase flow [12]; in a glass inner layer pipe [13]; in cooled heat exchanger tubes [14]; determining wax deposition using heat transfer method [15] and solvent migration in a paraffin deposit [16]. Examples of studies in mitigation of wax deposition can be found, e.g. using high velocity flow [2]; using the combined shear and flow improvers [17]; inductive heating of pipelines [18] application of glass inner layer [13] addition of chemical wax inhibitor [19], [20] using Nd–Fe–B magnets [21], [45] and use of exothermic chemical reactions [22]. In this work, we introduce a relatively new device, the oscillatory baffled tube apparatus (OBTA), which is completely different from these aforementioned.

OBTA is a mixing technology and offers more uniform mixing and particle suspension than traditional reactors. The fluid dynamic conditions in an OBTA are governed by two dimensionless groups, namely, the oscillatory Reynolds number (Reo) and the Strouhal number (St), defined asReo=2πfxoρLDcμSt=Dc4πxowhere ρL is the fluid density (kg m−3), μ the fluid viscosity (kg m−1 s−1), Dc the column diameter (m), xo the oscillation amplitude (m) and f is the oscillation frequency (Hz). The oscillatory Reynolds number describes the intensity of oscillation applied to the system, using the peak velocity of oscillation as the characteristic velocity and the column diameter as the characteristic dimension. The Strouhal number is inversely proportional to the amplitude of oscillation and represents the ratio of column diameter to stroke length, a measure of the effective eddy propagation. The product of 2πfxo is the maximum oscillatory velocity (m/s). This study has been prompted by the idea that the enhanced mixing in OBTA could lead to increase suspension of wax particles (with liquid) thereby reducing either the total deposition or the deposition rate. In this paper, an experimental investigation into wax deposition in the OBTA is reported, focusing on the effects of the paraffin wax content with changing oscillation frequencies and amplitudes on the percentage of wax deposition.

Section snippets

Mechanism

The mechanism of wax deposition in oil pipeline is largely via nucleation and crystallisation processes where oil is entrapped between crystals leading to gel formation and can be treated as bulk crystallisation [11]. Consequently, the Avrami phase transition equation is the well-known principle in describing such crystallisation kinetics. The original derivations by Avrami [23] have been simplified by Evans [24] and put into polymer context by Meares [25] and Hay [26]. The basic principle can

Materials and methods

The schematic diagram of the batch OBTA used in this study is shown in Fig. 1. The OBTA was made of a jacketed glass column of 25 mm in internal diameter, 50 mm in jacket diameter and 130 mm in height, giving a total liquid capacity of 64 ml and a working fluid capacity of 60 ml.

A K-type thermocouple is located within the OBTA and connected to a computer via Multipurpose Lab Interface (MPLI) supplied by Vernier Software Inc., USA, to record temperature profiles inside the column. A set of two

Results and discussion

In order to assess the effect of baffle oscillation on the wax deposition, control experiments were carried out in the OBTA with the presence of baffles, but without oscillation. Observations from the experiments were that the wax deposit was in the form of an oil gel, and the hardness of the gel increased with the content of the paraffin wax. The gel-like wax was adhered to the baffle surfaces and the wall of the column, while the liquid that flowed out from the centre of the column was a

Kinetic analysis

From the wax deposition profiles presented earlier, wax formation can generally be divided into two main sections: the growth phase and the quasi-steady (or the asymptotic) state. Typically the growth phase took place in the first 2 min of the experiments for all the wax solutions. By analysing the growth phase curves using the Avrami theory, some crystallisation/deposition kinetics can be extracted. Fig. 5, Fig. 6 plot log[−ln(1  δr)] versus log(t) for the effect of oscillation frequency and

Conclusion

The study shows that the oscillatory motion has two opposite effects on wax deposition within the OBTA: at low wax concentration, the oscillatory motion significantly reduced the percentage of wax deposition, e.g. 40–60% without the use of any solvent or wax inhibitor; and completely prevented 100% wax deposition from happening – the beneficial effect; at higher wax contents, on the other hand, the introduction of oscillatory motion not only promoted wax deposition, but also speeded up the

Acknowledgements

LI wishes to acknowledge Universiti Teknologi PETRONAS, Tronoh, Malaysia, for sponsoring his PhD programme and Heriot-Watt University, Edinburgh, Scotland for the placement and research facilities.

References (47)

  • A.J. Cordoba et al.

    Solvent migration in a paraffin deposit

    Fuel

    (2001)
  • I.M. El-Gamal

    Combined effects of shear and flow improvers: the optimum solution for handling waxy crudes below pour point

    Colloid. Surf. A

    (1998)
  • E. Marie et al.

    Settling of paraffin crystals in cooled middle distillates fuels

    J. Colloid Interface Sci.

    (2004)
  • B.F. Towler et al.

    Mitigation of paraffin wax deposition in cretaceous crude oils of wyoming

    J. Petrol. Sci. Eng.

    (2004)
  • C. Caze et al.

    A new method to determine the Avrami exponent by D.S.C. studies of non-isothermal crystallisation from the molten state

    Polymer

    (1997)
  • M.G. Lu et al.

    Curing behaviour of unsaturated polyester system analysed by Avrami equation

    Thermochim. Acta

    (1998)
  • A.R. Bhattacharyya et al.

    Crystallisation and orientation studies in polypropylene/single wall carbon nanotube composite

    Polymer

    (2003)
  • R.J. Campos et al.

    Fractionation of milk fat by short-path distillation

    J. Dairy Sci.

    (2003)
  • D.Z. Chen et al.

    Cure kinetics of epoxy-based nanocomposites analysed by Avrami theory of phase change

    Polym. Test.

    (2003)
  • W. Li et al.

    Isothermal crystallisation kinetics of poly(ethylene terephthalate)–poly(ethylene oxide) segmented copolymer with two crystallising blocks

    Polymer

    (2005)
  • S. Pal et al.

    Cocrystallisation mechanism of poly(3-alkyl thiophenes) with different alkyl chain length

    Polymer

    (2005)
  • R.P. Pereira et al.

    Nanostructure and crystallisation kinetics of poly(ethylene oxide)/poly(4-vinylphenol-co-2-hydroxyethyl methacrylate) blends

    Polymer

    (2005)
  • X. Wang et al.

    Crystallization kinetics of amorphous NiTi shape memory alloy thin films

    Scripta Mater.

    (2006)
  • Cited by (0)

    View full text