Meng Xu ,Xingyu Fng ,Fulong Ning ,*,Wenji Ou ,Ling Zhng ,Dongdong Wng

a Faculty of Engineering,China University of Geosciences,Wuhan,Hubei,430074,China

b Anhui Nuclear Exploration Technology Central Institute,Wuhu,241000,China

c National Center for International Research on Deep Earth Drilling and Resource Development,China University of Geosciences,Wuhan,Hubei,430074,China

ABSTRACT Natural gas hydrates are mostly formed in low-permeability and fractured muddy sedimentary formations.Adding suitable nanoparticles to the drilling fluid system can improve its filtrate resistance and fracture plugging,and effectively weaken the invasion of drilling fluid into the reservoir.However,it is likely that nanoparticles promote hydrate formation and accumulation in wellbores which will induce accidents.Therefore,this study investigated the effect of hydrophilic silica nanoparticles with particle sizes of 30 nm,60 nm,and 80 nm and concentrations of 0.5-4.0 wt% on hydrate formation during upward migration of methane gas using a dynamic simulation system for hydrate formation in a wellbore.The experimental results show that under the condition of methane gas migration,hydrophilic silica nanoparticles inhibit hydrate formation.The inhibition effect increased with the growth in the particle size under a constant concentration,whereas it first increased and then decreased with increasing nanoparticle concentration under a constant particle size.The strongest inhibition effect was observed at a hydrophilic silica nanoparticle concentration of 2.0 wt%.The influence of hydrophilic silica nanoparticles on hydrate formation may be mainly determined by their hydrophilic properties,heat and mass transfer,and gas migration in the wellbore.Our research indicates that hydrophilic silica nanoparticles can be added to hydrate drilling fluid systems if their concentration can be properly controlled.

Keywords:Natural gas hydrate Drilling fluid Hydrophilic silica nanoparticles Methane gas migration

1.Introduction

Gas hydrates are non-stoichiometric cage clathrates formed by host water molecules and guest gas molecules(mainly CH4)under high-pressure and low-temperature conditions.Gas hydrates are commonly known as“combustible ice”because they look like ice and are combustible in the case of fire[1].Gas hydrates are widely distributed in seabed sediments and permafrost zones.Because of their abundant reserves,many countries regard gas hydrates as a new alternative energy source for the future[2].The hydrate technique can also be used for gas storage and transportation[3],water desalination[4,5],and capture and sequestration of carbon dioxide[6,7].

There are still many challenges in the development and utilization of hydrates,among which safety risk is one of the greatest challenges[8].Most hydrates are distributed in low permeability,weakly consolidated,and fractured muddy sediments.Overbalanced drilling techniques are adopted to maintain wellbore stability and ensure wellbore production safety[9].However,in this case,the drilling fluid may invade hydrate formation and induce hydrate decomposition,causing borehole wall instability and log data distortion.At the same time,the decomposed water and gas may form hydrates again during drilling fluid circulation,causing safety accidents in the well[10-12].Therefore,reducing the damage of drilling fluid invasion into hydrate reservoirs is one of the problems that should be addressed in hydrate exploration and development.In recent years,nanoparticle materials have been widely used in conventional oil and gas drilling,unconventional shale gas drilling,and coal bed methane drilling[13,14].The advantages of nanoparticles are their small size,large specific surface area,and high surface activity.Nanoparticles cannot only fill microcrack channels but can also combine with high-molecular polymers to fill large pores and cracks by bridging and crimping[15].The addition of nanoparticles to drilling fluid can improve the filtration resistance,fracture sealing,and viscosity of the drilling fluid,as well as the quality of mud cake formed by the drilling fluid[16,17].Moreover,The addition of nanoparticles can also maintain wellbore stability and reduce the invasion of drilling fluid into a formation[18].Therefore,nanoparticles are very suitable for drilling hydrate formations with low permeability and weak consolidation.

However,existing research has shown that the addition of nanoparticles to the solution will have a certain influence on hydrate formation,and may promote the growth and aggregation of hydrates,which is very unfavorable in terms of the flow safety of wellbore drilling fluid[19-23].The existence of nanoparticles can reduce the surface tension of the contact surface between methane and water,increase the gas-liquid contact area of the system,and enhance the mass transfer process.Meanwhile,nanoparticles can effectively accelerate the heat transfer efficiency of the system,thus promoting hydrate formation[24-27].However,it has also been found that nanomaterials,especially some oxide nanoparticles,cannot promote hydrate formation and may even inhibit hydrate formation in some cases.Liu et al.[28]found that compared with pure water,nano-CuO and nano-SiO2prolonged the induction time of CO2gas hydrate formation.Nesterove et al.[29]reported that Al2O3fluid had both inhibiting and promoting effects on the induction time of hydrate formation.This difference was mainly due to the different sizes of the nanoparticles and different surface functional groups.Gas hydrate nucleation and growth are closely related to the hydrophobicity of the gas,and nanoparticles can either be hydrophilic or hydrophobic.Most studies have found that hydrophobic nanoparticles can promote the formation of hydrates[30,31].The influence of hydrophilic nanoparticles on hydrate formation has been less studied.Wang et al.[32-34]found that hydrophilic nano-SiO2promoted hydrate formation at some particle sizes and concentrations,whereas it inhibited hydrate formation at other particle sizes and concentrations.Therefore,the hydrophilicity,hydrophobicity,particle size,and concentration of nanoparticle may have different effects on hydrate formation.

Before using nanoparticles as additives in marine hydrate drilling fluid,it is necessary to clarify the effect of nanoparticles on hydrate formation in the wellbore fluid.Wang et al.[32-34]suggested that hydrophilic nano-SiO2could be added to the hydrate drilling fluid system because it can inhibit hydrate formation under certain conditions.However,their research was conducted in a static environment,in which the experimental conditions differed to the dynamic conditions of the actual wellbore environment.On this basis,considering the dynamic migration of shallow gas and hydrate decomposition gas in the wellbore during the actual hydrate drilling process,the influence of hydrophilic nano-SiO2on hydrate formation in the wellbore fluid under the condition of methane gas migration was studied using a self-developed dynamic hydrate formation simulation system in the wellbore.The mechanism of the influence of hydrophilic nano-SiO2on hydrate formation and accumulation was also considered.This study provides a reference for optimizing the type of nanoparticles and the design of hydrate drilling fluid systems,which are of great significance for ensuring the safety of hydrate drilling and protecting the hydrate reservoir.

2.Experimental

2.1.Experimental equipment

This study used a self-developed dynamic hydrate formation simulation system in a wellbore.Fig.1 presents a schematic of the experimental device,which mainly comprised a high-pressure visualized reactor,a temperature control system,a pressure control system,and a data acquisition system.The reactor was equipped with a baffle(the middle position of the high-pressure visualized reactor,as shown in Fig.1).The total overflow area of the baffle was approximately 0.5 cm2,and the area of a single circular hole was approximately 2.5 × 10-2cm2.The baffle was installed to simulate the relevant equipment or channels in the wellbore that impede gas migration,for example,the location of the borehole diameter change.To better simulate the rotation of the drill pipe in actual drilling/production and accelerate hydrate formation,an MS6-Pro digital magnetic agitator(Dalong Xingchuang Test Instrument(Beijing)Co.,Ltd.)with a rotation speed of 200-1500 rpm was installed at the bottom of the reactor.The DC-3015 temperature control system was a constant low-temperature reaction bath produced by Jiangsu Tianling Instrument Co.,Ltd.The pressure control system was composed of a pressure-regulating valve,back-pressure valve 1,and back-pressure valve 2.The gas injection rate was precisely controlled by adjusting the pressure-regulating valve and backpressure valve 1.A D07-11C gas mass flowmeter produced by Beijing Seven Star Huachuang Electronics Co.,Ltd.was used to monitor the gas injection rate.Back-pressure valve 2 was used to control the pressure in the reactor.This system injected gas into the reactor by means of constant-speed“bubbling,”and was equipped with a baffle and magnetic stirrers to better simulate the drilling bit rotation and gas upward migration in the actual wellbore.

2.2.Experimental materials

The methane gas used in this experiment had a purity of ＞99.9%and was supplied by Wuhan Newruide Special Gas Co.,Ltd.Deionized water was manufactured using LD-DI-Micro-20 deionized water equipment purchased from the Shanghai Liding Water Treatment Equipment Co.,Ltd.Hydrophilic nano-SiO2with particle sizes of 30 nm,60 nm,and 80 nm was supplied by Beijing Deco Island Gold Technology Co.,Ltd.,and was prepared by the gas phase method and carried out by surface modification.These particle sizes are the accurate sizes of nano-SiO2produced by the gas phase method,and are frequently used in most fields at present.

2.3.Experimental procedure

Conventional and unconventional oil and gas drilling fluids only require a small addition amount of solid-phase nanomaterials,which can significantly improve or affect many properties of drilling fluid systems[13].In this experiment,the nano-SiO2concentration was controlled in the range of 0.5-4.0wt%.The experimental temperature was controlled at 3°C,and the corresponding equilibrium pressure of methane hydrate was approximately 3.5 MPa[35].Considering that the rotating speed of the drill pipe in the drilling equipment was up to 600 rpm,the rotating speed of the magnetic stirrer in this experiment was set to 500 rpm.Each group of experiments was repeated three times to ensure the accuracy of the experiment.The experimental steps were as follows.(1)The reactor was cleaned with distilled water,and the air tightness was checked.(2)A high-speed stirring device(20 000 rpm)was used to configure the nanofluids.The top of the high-pressure reactor was opened,and 300 mL of nanofluid solution(or deionized water)was poured into it.Then,the top of the reactor was covered and sealed.(3)Methane gas was slowly and continuously injected for a period,and the residual air in the reactor was removed to ensure the accuracy of the experiment.(4)The magnetic agitator was turned on,and the speed was set to 500 rpm.(5)The temperature control system was opened,and the gas was simultaneously cooled in the buffer tank and liquid in the reactor to stabilize the temperature of the system at 3°C.(6)The pressure of back-pressure valve 2 was adjusted to 8 MPa(in all experiments,hydrate was formed before the pressure in the reactor reached 8 MPa).The buffer tank was opened,and methane gas was injected into the reactor at a steady flow rate of 300 standard cubic centimeters per minute(SCCM).If operating and observing are conducted simultaneously,it can be modified as:by observing the flow meter and operating the pressure regulating valve and backpressure valve 1.When the pressure in the reactor reached 3 MPa,the timing was started,and the phenomena in the reactor were observed.(7)When hydrate formation was observed in the reactor,the time of formation and the pressure at this time in the reactor were recorded.Methane gas was continuously injected into the reactor at a steady flow rate of 300 SCCM for 6 min,and the pressure in the reactor after 6 min was recorded.(8)After the experiment,the magnetic stirrer was closed and back-pressure valve 2 was adjusted to release the gas in the reactor.Then,the bottom of the reactor was opened to drain the solution and clean the device.

2.4.Data processing

2.4.1.Induction time

The induction time of hydrate formation is an important index for evaluating the promotion or inhibition of hydrate formation.At present,there is no accurate definition of induction time,which can be considered from the perspectives of theory and experiment.From a theoretical perspective,the induction time is the time from the stable state of the system to the occurrence of a stable critical nucleus[36,37];however,this is difficult to measure experimentally.From an experimental perspective,the induction time is the time from the initial stable state of the system to the first detectable hydrate formation of crystals[38],which can be easily obtained by experimental measurements.In this study,the second method was used to determine the induction time for hydrate formation.The temperature and stirring speed of all experiments were the same,and methane gas was uniformly injected at the same“bubbling”rate.The initial stable state of the system was regarded as that when the pressure in the reactor reached 3 MPa.Subsequently,methane gas was injected at the same rate.When hydrate crystals were observed,the induction time of hydrate formation and the pressure in the reactor were recorded.The effect of hydrophilic nanoparticles on hydrate formation was evaluated by comparing the induction time and pressure in reactors with different particle sizes and concentrations of hydrophilic nanoparticles and deionized water.

2.4.2.Average rate of hydrate formation

The average rate of hydrate formation is another important index for evaluating the promotion or inhibition of hydrate formation,which can be calculated using experimental temperature and pressure data.As shown in Fig.2,the red line indicates the pressure growth mode in the reactor when methane gas was injected into the reactor at a constant speed without hydrate formation during the experiment(by turning off the magnetic stirrer to prolong the induction time of hydrate formation,without hydrate formation within the required time range).The green line represents the pressure growth mode in the reactor after hydrate formation.When hydrate formation was observed,methane gas continued to be uniformly injected at the same“bubbling”rate.In the experiment,back-pressure valve 2 was set to 8 MPa to ensure that only methane gas was injected,and no methane gas was discharged during the entire experiment.In this process,hydrate was formed continuously,and thus,methane gas should have been consumed continuously.The pressure was reduced relative to the red line,without hydrate formation.The pressure decreased(ΔP,MPa)during Δt(min),as shown in Fig.2.Because the gas used in the experiment was high-purity methane(≥99.9%),the amount of hydrate formation could be characterized by the consumption of methane gas.Based on the calculation formula of gas consumption derived by Li et al.[39],and by neglecting the fact that hydrate formation will cause a certain volume expansion,combined with the characteristics of the experimental equipment used in this work,the calculation formula of gas consumption in the process of methane hydrate formation is derived as follows:

where Δnis the amount of methane consumption(mol);P1is the pressure(MPa)of the reactor after Δt(min)when there is no hydrate formation indicated by the red line in Fig.2,which is directly obtained from the red line;P2is the pressure(MPa)of the reactor after Δt(min)time when hydrate formation is observed,indicated by the green line in Fig.2;Vis the volume of methane in the reactor(m3),which was 200 mL in this work;Ris the gas constant(8.31441 J·mol-1·K-1);Tis the experimental temperature(K),which was 276.15 K in this work;Z1andZ2are the compressive factors under pressuresP1andP2,respectively,which can be calculated as follows[40]:

whereAiis a given parameter(A1=0.31506237,A2= -1.0467099,A3= -0.57832729,A4= 0.53530771,A5= -0.61232032,A6=-0.10488813,A7=0.68157001,andA8=0.68446549)[41];Pris the comparative pressure;Tris the comparative temperature;and ρris the methane comparative density.Pr,Tr,and ρrwere calculated as follows:

wherePis the methane gas pressure(MPa),Pcis the critical pressure of methane gas(～4.60 MPa)[39];Tis the methane temperature(K);andTcis the critical temperature of methane gas(～191.11 K)[39].

When hydrate formation is observed,hydrates will continue to grow in a reactor.Previous studies have shown that the growth rate of hydrates is faster in the initial period(Δt),but slower in the later period[42];therefore,the growth rate of hydrates in the initial period is of great significance in this work.The value of Δtin this study was 6 min.The hydrate formation rate obtained by calculation refers to the average rate of hydrate formation from the beginning of formation to continuous growth for 6 min.The calculation is derived from Eq.(1),as follows:

wherevis the average consumption rate of methane gas(mol/min).

3.Results

3.1.Hydrate formation in deionized water without nanoparticles

To compare and analyze the effect of nanoparticles,a simulation experiment of methane hydrate formation in deionized water was carried out.At the beginning of the experiment,methane gas uniformly entered the reactor from the bottom inlet of the reactor.Bubbles were produced and migrated upward in the liquid until the bubbles broke and disappeared at the gas-liquid interface.The baffle produced some resistance to the migration of bubbles;thus,bubble aggregation was observed at the lower part of the baffle.As time progressed,the pressure in the reactor increased and many white fine hydrate crystals were observed near the gas-liquid interface.Then,the glass wall of the visual window close to the upper part of the gas-liquid interface became“foggy”and fuzzy.At the same time,the hydrate membrane was found on the surface of bubbles at the gas-liquid interface and at the lower part of the baffle(Fig.3a and b).Fig.3a shows the entire gas-liquid interface overlooked from the visual window located at the upper part of the gas-liquid interface.Bubbles arriving at the gas-liquid interface no longer broke and disappeared,but formed hydrate membranes that rapidly and continuously aggregated on their surfaces.The bubbles that were“intercepted”at the baffle also rapidly formed hydrate membranes that aggregated on their surfaces.The rate and amount of aggregation at the baffle were much higher than those at the gas-liquid interface,as shown in Fig.3c.

The initial formation of hydrates at the gas-liquid interface and in the lower part of the baffle was consistent with the hypothesis of hydrate nucleation at the interface[37,43].The gas-liquid contact area was large in these two places,providing more nucleation points for hydrate formation.At the same time,the baffle blocked the flow of methane gas,resulting in a relatively high gas concentration.Therefore,at the lower part of the baffle,the aggregation velocity and number of bubbles that formed hydrate membranes on their surfaces were considerably higher than those at the gas-liquid interface.After hydrate formation,the new bubbles continued to migrate upward to the baffle and collided with the bubbles wrapped by the hydrate membranes.

There is an adhesion force between the bubble wall composed of water molecules and the hydrate membrane,such that bubbles are easily retained and quickly form hydrate membranes on their surfaces.They then become the media for the next bubble to be“retained”[44-46].Very few bubbles get rid of the adhesion force between the bubble wall and the hydrate membrane and the resistance of the baffle.They continue to migrate upward to the gas-liquid interface to remain and form hydrate membranes.The average induction time and pressure of hydrate formation in deionized water were 2239 s and 6.032 MPa,respectively,and the average gas consumption rate was 1.549 × 10-3mol ·min-1.

3.2.Hydrate formation in nanoparticle solutions

In this study,hydrophilic nano-SiO2with particle sizes of 30 nm,60 nm,and 80 nm and concentrations of 0.5-4.0 wt%were selected for the experiment.The average values of the parameters for each set of experiments are listed in Table 1.The effect of hydrophilic nano-SiO2on methane hydrate formation was assessed by calculating two kinetic indices:the induction time and average rate of hydrate formation(Table 1).

Table 1 Hydrate formation in deionized water with/without hydrophilic nano-SiO2 at 3.0 °C.

3.2.1.Induction time

It was difficult to observe the upward migration of gas bubbles because the solution of nanoparticles was not completely transparent when methane gas entered the reactor uniformly at a certain flow rate.After a certain period,as in the pure water system,fine white hydrate crystals were observed floating at the gas-liquid interface.Then,the glass wall of the visual window close to the upper part of the gas-liquid interface became“foggy”and fuzzy.It is speculated that there may have been some residual nanoparticle solution on the glass wall due to the stirring of the magnetic agitator,and that hydrate was then formed on the glass wall near the gas-liquid interface.It was also observed that the bubbles no longer broke at the gas-liquid interface,but rapidly formed hydrate membranes on their surfaces.This can be seen in Fig.4a-c,which shows the entire gas-liquid interface overlooked from the visual window located at the upper part of the gas-liquid interface.

Fig.5 shows the induction time in each set of experiments.Except for the hydrophilic nano-SiO2concentrations of 0.5 wt%and 4.0 wt%,all other concentrations at different particle sizes exhibited an inhibition effect on hydrate nucleation,but to different degrees.For the three particle sizes of hydrophilic nano-SiO2,the changes in the induction times for hydrate formation as a function of concentration exhibited similar trends.At nanoparticle concentrations of 0.5 wt%and 4.0 wt%,the induction times were shorter than that of deionized water,indicating that nano-SiO2at these concentrations promoted the nucleation of hydrate in the fluids.In contrast,at nano-SiO2concentrations of 1.0-3.5 wt%,the induction times were longer than that of deionized water,indicating that nano-SiO2at these concentrations had an inhibition effect on hydrate nucleation in the fluids.The induction times gradually increased and the inhibition effect gradually strengthened when the concentrations ranged from 1.0 wt% to 2.0 wt%.At a nano-SiO2concentration of 2.0 wt%,the induction times were 2624 s,2668 s,and 2745 s for the particle sizes of 30 nm,60 nm,and 80 nm respectively,thus presenting the strongest inhibition effect on hydrate nucleation.With an increase in the nano-SiO2concentration,the induction times gradually decreased and the inhibition effect gradually weakened.The effect changed from inhibition to promotion when the concentration reached 4.0 wt%.Fig.5 also shows that the inhibition effect on hydrate nucleation strengthened with an increase in the particle size of hydrophilic nano-SiO2.

3.2.2.Average rate of hydrate growth

After some time,even if the solution was not transparent,hydrates could be observed at the lower part of the baffle due to the growth and aggregation of hydrates.Fig.6 illustrates the growth and aggregation of hydrates at the baffle of nanoparticle solutions with a particle size of 60 nm and different concentrations.The growth and aggregation of hydrates were similar at particle sizes of 30 nm,60 nm,and 80 nm.

Fig.7 plots the average gas consumption rate within 6 min after hydrate formation in each set of experiments.Hydrophilic nano-SiO2with particle sizes of 30 nm,60 nm,and 80 nm slowed down the rate of hydrate growth.The changes in the average gas consumption rate as a function of concentration exhibited similar trends for the three particle sizes of hydrophilic nano-SiO2.At nano-SiO2concentrations of 0.5 wt% and 4.0 wt%,the average gas consumption rates were faster than that of deionized water,indicating that nano-SiO2accelerates the rate of hydrate growth.In contrast,at nano-SiO2concentrations of 1.0-3.5 wt%,the average gas consumption rates were slower than that of deionized water,indicating that nano-SiO2has a slowing down effect on hydrate growth.The average gas consumption rate gradually decreased and the slowing down effect gradually strengthened with the increase in the nano-SiO2concentration from 1.0 wt%to 2.0 wt%.At a nano-SiO2concentration of 2.0 wt%,the average gas consumption rates were 0.605 × 10-3mol·min-1,0.489 × 10-3mol·min-1,and 0.392×10-3mol·min-1for the particle sizes of 30 nm,60 nm,and 80 nm,respectively,thus corresponding to the strongest slowing down effect on hydrate growth.Then,with the increase in the nano-SiO2concentration,the average gas consumption rate gradually increased and the slowing down effect of nano-SiO2on hydrate growth gradually weakened.In addition,the effect changed from decelerating to accelerating when the concentration reached 4.0 wt%.Fig.7 also shows that the slowing down effect on hydrate growth strengthened with an increase in the particle size of hydrophilic nano-SiO2.

4.Discussion

The experimental results revealed that hydrophilic nano-SiO2promoted the formation of hydrate at concentrations of 0.5 wt%and 4 wt%,but inhibited the formation of hydrate at concentrations of 1-3.5 wt%.This behavior was mainly determined based on the hydrophilic properties of hydrophilic nano-SiO2,the enhancement of heat and mass transfer of the system,and the migration of methane gas in the wellbore.

4.1.Hydrophilic properties of hydrophilic nano-SiO2

Due to the existence of a certain amount of hydrophilic hydroxyl groups on the surfaces of hydrophilic nano-SiO2particles,the water molecules nearby will be bound to some extent,and some water molecules will be wetted and adsorbed on the surfaces of nanoparticles to form a“water film.”Under the action of this mechanism,water molecules are likely to exist as free water molecules and bound water molecules in the fluid prepared by hydrophilic nano-SiO2and deionized water,as shown in Fig.8.The existence of bound water molecules in nanofluids reduces the number of free water molecules and the activity of water molecules,thus inhibiting hydrate formation.At the same time,only a small amount of hydrophobic non-polar CH4molecules can be distributed on the surfaces of hydrophilic nano-SiO2particles owing to their unique characteristics.Therefore,only a small amount of bound water molecules may participate in the formation of hydrates,which also plays a role in delaying the induction time and rate of hydrate formation[32-34].

Figs.5 and 7 show that the inhibition ability of hydrophilic nano-SiO2on methane hydrate formation increased with an increase in particle size.This may have been because the hydrophilicity of hydrophilic nano-SiO2increases with an increased particle size.Nanoparticles exist as“aggregates”in fluid.Due to the strong surface properties of nanoparticles,the larger the particle size,the weaker the aggregation of nanoparticles in fluid,and the smaller the size of aggregates.The smaller the size of the aggregates formed by nanoparticles,the larger the specific surface area of the aggregates,resulting in an increase in the number of bound water molecules and a decrease in the number of free water molecules in the fluid[32-34].Therefore,under the same concentration conditions,the inhibition of hydrate formation by hydrophilic nano-SiO2can be ranked by particle size as:80 nm ＞60 nm ＞30 nm.

4.2.Heat and mass transfer of hydrophilic nano-SiO2

The formation of hydrates is also controlled by heat and mass transfer[47].From the perspective of heat transfer,like a solid particle,the thermal conductivity of hydrophilic nano-SiO2is approximately two orders of magnitude larger than that of pure liquid.The addition of nanoparticles to deionized water enhances the heat transfer capacity of the system and facilitates the transfer of heat released during hydrate formation.The thermal conductivity of a hydrophilic nano-SiO2solution increases with an increase in concentration[32,47,48].From the perspective of mass transfer,the intense Brownian motion of nanoparticles in liquid reduces the surface tension of the gas-liquid interface and increases the gas-liquid contact area of the system.At the same time,the existence of nanoparticles provides many nucleation points for hydrate formation,thus promoting the hydrate formation.With an increase in concentration,more“nucleation points”act as hydrate nucleation and the mass transfer of the system strengthens[33,49].Therefore,the hydrophilic property of hydrophilic nano-SiO2can inhibit hydrate formation,while an enhanced heat and mass transfer of the system can promote hydrate formation.In the case of different concentrations,a certain factor plays a leading role.

4.3.Influence of the hydrophilic nano-SiO2 concentration on hydrate formation

When the hydrophilic nano-SiO2concentration was 0.5 wt%,the addition amount of nanoparticles was relatively low,and there were more unbound water molecules in the system than at higher concentrations.At this time,nanoparticles played a leading role in increasing the gas-liquid contact area,providing nucleation points for hydrate formation and enhancing the heat conduction capacity of the fluid,which shortened the induction time of hydrate formation and accelerated the rate of hydrate growth.With an increase in the nanoparticle concentration,the number of bound water molecules in the system increased,and the free water molecules decreased gradually.The inhibition effect of hydrophilic nano-SiO2on hydrate formation was highlighted,and hydrate formation was inhibited.When the concentration increased to 2.0 wt%,the ability of hydrophilic nano-SiO2particles to inhibit hydrate formation peaked.As the nanoparticle concentration continued to increase,the ability of nanoparticles to inhibit hydrate formation decreased,and even changed to a promoting effect.Three reasons may explain this behavior.First,with an increase in the nanoparticle concentration,the size of the aggregates in the fluid increased further,resulting in the weakening of the strong surface properties of the nanoparticles and the binding effect on water molecules[32-34].Second,the collision between nanoparticles or aggregates intensified with the stirring action when the nanoparticle concentration reached a certain value,thereby creating a stronger destructive effect on the water film adsorbed on the surfaces of nanoparticles and producing more free water molecules.Third,with the increase in the nanoparticle concentration,the heat and mass transfer effects of the system were also enhanced.Fig.9 illustrates the effect of different concentrations and particle sizes of nano-SiO2on hydrate formation.

4.4.Influence of methane gas migration in the wellbore

The results of this study are similar to those obtained by Wang et al.[32-34].That is,with the increase in the hydrophilic nano-SiO2concentration,the corresponding inhibition effect on hydrate formation first increased before decreasing,and the inhibition effect was strongest at a certain critical value.However,the critical concentration obtained in this work was 2 wt%,whereas that obtained by Wang et al.was 4 wt%[32-34].This difference may be attributable to the continuous migration of methane gas in the wellbore.From the above analysis,it can be seen that nanoparticles have three influences on hydrate formation.First,nanoparticles can bind with water molecules and convert free water molecules into bound water molecules,thus inhibiting the formation of hydrates.Second,nanoparticles can promote heat and mass transfer,thus promoting hydrate growth.The occurrence of inhibition or promotion depends on the nanoparticle concentration.In this experiment,the stirring of the high-speed rotor at the bottom of the reactor was beneficial for improving the methane gas concentration in the solution and the disturbance degree of the entire system,thus promoting hydrate formation.Moreover,methane bubbles continued to migrate and were blocked by the baffles in the middle of the reactor,resulting in an increase in the gas-liquid contact area and gas concentration,which were conducive to hydrate formation.When bubbles moved upward again through the baffle and broke at the gas-liquid interface,the water surface fluctuated and carried a small amount of liquid to“break out”in the gas phase space,as shown in Fig.10.This small volume of liquid surrounded by gas was more conducive to hydrate formation.During the experiment,bubbles gathered at the lower part of the baffle and formed a hydrate film at the bubble boundary for accumulation.Hydrate particles were initially formed at the upper gas-liquid interface,and then bubbles formed a hydrate film and gradually accumulated(Fig.3).Both phenomena indicate that hydrate was more easily generated under this experimental condition than in the static system of Wang et al.[32-34].Hydrate formation was promoted by bubbling and baffle blocking.Therefore,compared with the conditions without bubbles and baffles,the inhibition effect of nanoparticles(at the same concentration)on hydrate weakened,resulting in a decrease in the optimal inhibitory concentration of nanoparticles in this study.This indicates that the selection of the nanoparticle concentration for the drilling fluid is very important.Moreover,hydrates are easily generated at the gas-liquid interface.Therefore,in drilling or production processes,gas and liquid flow,especially in locations where free gas and free water can contact or migrate(e.g.,the gas-liquid separator and screen pipe),will have a great risk of blockage caused by secondary hydrate generation.

5.Conclusions

In this study,experiments on hydrate formation in deionized water with and without hydrophilic nano-SiO2were conducted under the conditions of methane gas migration using a dynamic simulation system for hydrate formation in a wellbore.By comparing the induction time and average rate of hydrate growth between the hydrophilic nano-SiO2solution and deionized water,the influence of hydrophilic nanoparticles with particle sizes of 30 nm,60 nm,and 80 nm and concentrations of 0.5-4.0 wt%on hydrate formation was determined.The results show that all nano-SiO2concentrations except 0.5 wt% and 4.0 wt% exhibit different degrees of inhibition effect on hydrate formation.The inhibition effect first increased and then decreased with an increase in the nano-SiO2concentration under a constant particle size,and strongest inhibition effect was observed at a concentration of 2.0 wt%.Under a constant concentration,the inhibition effect increased with an increase in the particle size.The critical value of the hydrophilic nano-SiO2concentration to achieve the strongest inhibition effect was reduced when methane gas was transported in the wellbore.

Our study suggests that hydrophilic nanoparticles can be added to drilling fluid system as they exhibit a certain degree of inhibition on hydrate formation if their concentration can be controlled properly.Owing to the limitations of the experimental conditions,the influence of hydrophilic nano-SiO2particles on hydrate formation was studied from a macroscopic perspective.Our next work will use molecular dynamic simulations and other methods to study the influence mechanism at a deeper level to make the research results more valuable and applicable in practical engineering.

Declaration of competing interest

The manuscript is approved by all authors for publication.We authors confirm that no conflict of interest exits in the submission of this manuscript.The original work has not been published previously and is not under consideration for publication elsewhere,in whole or in part.

Acknowledgements

This work was supported by the National Natural Science Foundation of China(Grant No.41672367,51704266,and 51874263)and was partly supported by the National Key Research and Development Program of China(No.2018YFE0126400),and the Special Project for Marine Economic Development(Six Major Marine Industries)of Department of Natural Resources of Guangdong Province(GDNRC[2020]047),and the Fundamental Research Funds for National Universities,China University of Geosciences(Wuhan)(Grant No.CUGGC09).