Sand consolidation using enzyme-induced carbonate precipitation: new insights on temperature and particle size effects

Sand production is a major issue in the oil and gas industry. Unconsolidated sand can be produced with the oil or gas a cause many issues to the production facilities. Enzyme-induced carbonate precipitation (EICP) is a promising method for sand consolidation and is characterized by its environment friendliness. Numerous studies have shown its effectiveness in ambient conditions. However, oil and gas downhole well operations are high pressure and high-temperature conditions. The objective of this study is to investigate effect of high temperature on EICP reaction and its efficiency in terms of uniformity to consolidate different types of sand samples. In this paper, the behavior of EICP solutions is examined in high temperatures from 25 to 90 °C. The study shows that high temperature environment doesn’t handicap efficiency but in contrast it can favor the reaction if optimum concentration of reactants has been selected. The temperature effect is also discussed in terms of controllability of reaction which can favor application of reaction. Qualitive analysis shows when EICP solutions containing more than 50,000 ppm of metal ions and stoichiometrically surplus urea requires exposure to heat for reaction progress. The effect of sand particle size and its implication on the consolidation process was examined. Particle size of fine and medium sand ranged from 125 to 250 µm and 250 to 425 µm respectively while for coarse sand 70% sand particle size was between 425 and 700 µm. Designed EICP solutions achieve 9,000 psi for medium and almost 5,000 psi intrinsic specific energy for coarse sand samples. However, treated samples were subject to non-uniform distribution of strength of which can be up to 8,000 psi difference between top and bottom half of the samples.

where m p is the mass of precipitated material, m t is the theoretical mass, C is the concentration of solution (mol/L), V is the volume of solution and M is the molar mass of precipitant.
Nevertheless, temperature effect on EICP reaction is not studied by many authors.According to studies of Nemati et al. temperature increase from 20 to 50 °C effects positively on the final yield of reaction 5 .The favorable effect is also validated by Manar AlAhmari studies where maximum temperature is raised to even to 140 °C.It has been precipitation process happens after 100 °C regardless of presence of urease 4 .Another study on temperature effect has been done by Feder et al.Different types of catalyzers for urea hydrolysis has been tested and their performance monitored for different temperatures.Results show that temperature increment to 60 °C can stimulate urea hydrolysis while after that point enzymes (jack urease bean) can lose its functionality 2 .Many other additives for sand consolidation are studied in the literature such as different types of polymers and nanoparticles, each additive has its limitations when it comes to application in the oil and gas industry 22,23 .
Despite those studies, there is research gap in detailed analysis of temperature effect on EICP and their performance on consolidation process.In more detail, studies on EICP consolidation are limited to single strength measurement for treated samples.However, if the application is extended from laboratory scale to field scale, uniformity, and precipitation distribution will be one first factor to be considered.
The objective of study is to investigate effect of high temperature (higher than ambient condition) on EICP reaction and its efficiency in terms of uniformity to consolidate different types of sand samples.Their effect will be measured and analyzed through predesigned reaction and consolidation metrics.

Materials and equipment
Reactants and catalyzer.To perform temperature effect studies on EICP, reactants which included MgCl 2 *6H 2 O, CaCl 2 *2H 2 O, urea and catalyzer of urease from jack bean has been used.Those chemicals are product from Sigma Aldrich and VWR chemicals with 99% purity.
Enzymes are biopolymeric macromolecules and possess 3-dimensional structures with an ability to catalyze the reactions without being consumed.They achieve faster rate of reactions by lowering activation energy between reactants and transition state.The type of enzyme that has been used in this study is urease enzyme which is commonly occurring protein in higher order plants or microorganisms.Urease can be derived from different sources, but jack-bean urease is one of widely used and studied catalyzer 24 .Type of enzyme used in this study is Type III with 15,000-5,000 units/g solids.Enzymes catalyzer for this reaction is kept in refrigerator.Solubility of chemicals in solvent is validated using at least 10 min curing for each solution.
Urea, also known as carbamide, is an organic compound with chemical formula of CO(NH 2 ) 2 .The structure is formed by 2 amino groups and combined by carbonyl functional group.Other characteristics include good solubility in water, neither acidic nor basic and characterized by colorlessness and odorless solid state (powder).
Main source of carbonate ions is from hydrolysis of urea.A faster rate of hydrolysis is achieved by nucleophilic attack of urea carbonyl oxygen to nickel atoms of urease active site 25,26 .The interaction is enhanced by hydrogen bonding between amino groups of urea and Ni atom.Hydrolysis will initiate by resulting release of NH 3 which weakens urea and urease coordination and finally by breaking of carbamate group NH 2 COO -.This unstable group will give rise to formation of CO 2 .A summary of the materials used in the EICP solution and their functions is provided in Table 1.

Sand samples.
Prior to the treatment of sand, raw sand samples have been exposed to mineralogical analysis.With regard to the mineralogy (Fig. 1) where calcite and quartz are main constituents, the sand used in this study can be characterized as sandstone.There was also a small trace of clay minerals such as albite.To avoid their negative effect on consolidation, raw sand samples are exposed to water flushing and followed by filtering using vibratory sieve shaker.

Coupled plasma optical emission spectroscopy.
To measure metal ionic concentration of solutions, Inductively Coupled Plasma Optical Emission spectroscopy (Optima 8000 ICP-OES) has been utilized .The working principle of equipment is based on analytical analysis of emitted light from exposed solution.Each type of atom or ions carries unique characteristics of energy absorption and subsequent transition of energy level for their electrons.Energy level transition happens when electrons are exposed to high energy source (argon plasma).When excited electrons bounce back to their ground level, they will emit light with specific wavelength (

Sand sieving.
Prior to the treatment of sand by EICP solutions, it has been filtered using Retsch Vibratory Sieve Shaker.To achieve desired sand particle distribution, at least 10 min are given for equipment to vibrate the sand sample.Sand are sieved between 125 and 250 µm to achieve for finer ones while for medium sand between 250 and 425 µm sieves has been used.The particle size higher than 425 µm are considered as coarse sample.Three types of sand have been prepared and they are also flushed with water to avoid any contamination.All sand samples are oven dried at 90 °C for at least 4 days.
Scratcher test.The Wombat scratcher machine was used for the scratch test.This technique of assessing rock mechanical property is considered as quasi nondestructive test.It measures both tangential and normal components of force.Adjustable parameters are relative velocity between cutter and rock, depth of cut and cutting angle.
Based on geomechanical behavior of rock through, it will perform ductile or brittle behavior.Especially for ductile, there is a well-studied correlation between intrinsic specific energy and uniaxial compressive strength.Intrinsic specific energy is required energy to remove a unit volume of rock.This method offers several advantages over conventional methods such minimal sample preparation, non-destructive, reproducible tests, and ability to create log of strength through the sample 28,29 .This will allow for the recording of strength measurement with several data points along the entire length of the consolidated sample.

Methodology
Temperature sensitivity analysis.To perform temperature sensitivity analysis, reactant of reaction which are CaCl 2 *2H 2 O, MgCl 2 *6H 2 O and Urea have been prepared following the procedure.First, mix different concentration of CaCl 2 *2H 2 O, MgCl 2 *6H 2 O and Urea using mixer for at least one hour.Then, filter the solution.After that, cure the solutions for at least one day.Lastly, measure ionic concentration using Optima 8000 ICP-OES.The concentration of solutions is in increasing order from M4 to M2 and their concentration can be found in Table 2.For all solutions, molar ratio of urea to metal ions are taken higher than one to ensure favorable condition for precipitation such as high pH and adequate amount of carbonate ions.
Preparation of reactants has been followed by EICP solution preparation.In contrast to most studies, sensitivity analysis has been performed based on volume ratios.By keeping constant volume of 40 ml, the volume of catalyzer has been shrunk from 20 to 5 ml for each M2, M3, M4 reactants (Table 3).The general procedure starts by preparing 1 g/L solution from jack-bean urease.Then, cure catalyzer for 1 h and filter it.Then, prepare  To perform temperature sensitivity analysis, the temperature has ranged from room temperature to 90 °C with an increment of 10 °C.At each temperature increment, a 0.1 ml sample has been taken and diluted 1000 times using deionized water as described in Fig. 2. Those samples are exposed to elemental concentration measurements by Optima ICP-OES.At the final step, induced precipitation has been collected and dried for mineralogical analysis.
Consolidation studies.Sand preparation.Prior to the treatment of sand by EICP solutions, it has been filtered using Retsch Vibratory Sieve Shaker.To achieve desired sand particle distribution, at least 10 min are given for equipment to vibrate the sand sample.Sand are sieved between 125 and 250 µm to achieve for finer ones while for medium sand between 250 and 425 µm sieves has been used.The particle size higher than 425 µm are considered as coarse sample.Three types of sand have been prepared and they are also flushed with water to avoid any contamination.All sand samples are oven dried at 90 °C for at least 4 days.
Sand particle size distribution is shown in Fig. 3.It can be concluded that desired distribution has been achieved using sieve shaker with some minor population outside the range.For example, fine sand consists of 3% of particle which is lower than 125 µm while this is 5% of less than 250 µm for medium sand.3. Since the main target of study is to analyze sand particle size effect, maximum amount of catalyzer has been taken to ensure adequate amount of precipitation (Table 4).
After preparation of loose sand samples, the approximate pore volume of sand samples has been determined.The mass of samples ranged from 162 to 177 g and occupied a cell volume of 40 mm diameter and 60 mm height.All the samples are treated with 40 ml of EICP solution which is higher than pore volume to make sure full contact of sand.Mix-compact treatment method is applied, and samples are cured at least 10 days at 90 °C.

Results and discussion
Temperature sensitivity analysis.The progress of EICP reaction depends on concentration of metal ions and source of carbonate ions.Hydrolyzed urea enriches the solution with carbonate ions and when those ions confront with reactive metal ions such Ca 2+ and Mg 2+ , they give rise calcium carbonate, dolomite or other chemical compounds.Since precipitation process consume metal ion concentration of solution, consumption trend of Ca 2+ and Mg 2+ can help to track the progress of reaction.This method is more accurate and advantageous over conventional EICP sensitivity tests which is based on weight of precipitation.Considering the high cost of enzyme catalyzer, most of test-tube studies in literature are conducted in limited volumes (up to 50 ml) that yields little weight of precipitates which is difficult to analyze and choose optimum concentration.
It can be observed from Fig. 4 all of the solutions experience uninterrupted downward trend of Ca 2+ consumption as they are exposed to higher temperatures.Especially for highly saturated solutions M2 in Fig. 4a, there is still potential carbonate precipitation.M2V3 solutions which contain negligible amount of catalyzer volume to reactant exhibit competitive trend against M2V1 which has equal volume ratios of reactant and catalyzer.Regarding EICP solutions from less saturated M3 solution, effect of temperature become less pronounced.For example, M3V1 has achieved equilibria at 50 °C and incremental temperature is no longer effective for precipitation process.However, as in M2 solutions, higher temperature becomes necessity when volume of jack urease bean solution has been shrunk.As can be seen from Fig. 4b, increment after 50 °C leads to fluctuations in metal ion concentration probably due to generation of carbonic acid.However, this followed by equilibria shift to carbonate ions and consumption by metal ions for M3V2 and M3V3.A solution containing comparatively least amount of urea and metal salts shows interesting behavior in terms of efficiency of reaction.M4V1, as described in Fig. 4c, experiences sharpest and fastest drop of calcium concentration reaching a value of below 10,000 ppm.
Another interesting question is whether Mg 2+ is effective to get efficient precipitation in terms of mineralogy and mass.In spite of numerous literatures showing positive of effect of Mg 2+ presence for mass of precipitation 1,30 , they exhibit quite unreactive nature.From Fig. 5a to 5c, it is clearly observed that Mg 2+ consumption is significantly lower than Ca 2+ .In a high concentration environment such as in M2 and M3, there is lesser chance of contribution to precipitation while some contribution can be expected if solution contains less than those.
It has been experimentally validated that temperature fills the gap of catalyzer for induced carbonate precipitation in contrast to other studies which shows efficiency of enzyme degrades after 70 °C2 .However, trends in Figs. 4 and 5 show that the reaction progresses up to even 90 °C temperature.This fact can be explained by the nature of the reaction.Since urea hydrolysis is endothermic 31 , continuous supply of heat for more than 7 days stimulates the hydrolysis process and induces precipitation.
Since this progress is earned also at the expense of high reactant concentration, efficiency of reaction is another factor to consider.Figure 6 illustrates, total change of Ca 2+ concentration and estimated efficiency based    www.nature.com/scientificreports/ on consumed and introduced reactants.For combinations of M2 and M3, solutions containing more reactant to catalyzer induce more Ca 2+ consumption.In terms of efficiency, nearly half of existing ions lead to precipitation of carbonate minerals.Nevertheless, there is a clear sign of enhancement for M4V1 solution where almost 80% of existing Ca 2+ ions converted to precipitates.The faster and efficient reaction rate can be attributed better enzyme performance at low metal concentration 6 .Since contribution of Mg 2+ to precipitation is negligible according to ionic studies, the question arises on whether it has any effect on mineralogy.Figure 7 illustrates minerals comprising precipitation collected from test tubes.Percentage of dolomite mineral that is almost below 5% further confirms that Mg 2+ is less competent to Ca 2+ to induce carbonate precipitation in high concentration environment.On the other hand, minerals are not only characterized based on their chemical formula but also their crystal structure.For all solutions, a minimum of 40% of calcium carbonate is in the form aragonite morphology while the rest possesses calcite structure.From the aspect of soil improvement, it is debatable which one of them is more efficient.
Figure 7 also highlights the mass of precipitates.It fluctuates between 2.5 and 3.5 g.Solutions categorized by M2 and M4 show similar trends and weight of mass.Nevertheless, the mass of M3V2 which is almost 1 g higher than average mass of all is worthy to consider.This reaction product contains 50% less catalyzer than its closest counterpart and highlights how high temperature environment can stimulate reaction in the lack of catalyzer.
Deficit volume of enzyme is not only effective in terms of cost but also can be effective in terms of application.For ambient temperature, rate of enzyme induced precipitation is a function of strength of catalyzer and concentration.From the aspect of application which is not limited to fracture plugging and lost circulation mitigation, EICP cement need significant amount of time to imbibe through pore space and once stabilized to contact with formation 5,32 Click or tap here to enter text..When the reaction is too fast, precipitates will clog the pore space and will handicap further movement of fluid to the contact points of soil particles.This scenario is most probable for low reactant concentrations.The solutions containing 50% percent of catalyzer, especially M4V1, give rise to significant precipitation in short time.However, when catalyzer volume shrinks to 25% and even less 12.5%, the temperature and timing effect dominates.Those solutions can be easily delivered to the area of interest and precipitation can be induced by exposing them to higher temperature in comparatively longer period of time.Table 5 summarizes the visual inspection of EICP occurrence of all the samples at different temperatures from 25 to 90 °C.
Sand Consolidation Using EICP..The second part of the study highlights how effective the EICP solutions are to consolidate different types of sand.To measure strength distribution, the sample has been exposed to scratch test and at the end of each, high resolution camera is used to take picture of scratched samples.Prior to www.nature.com/scientificreports/distribution analysis of both strength and precipitation, it is worthy to understand how treatment methodology can affect the results.Sands treated with injection or surface percolation of EICP solutions are most vulnerable ones to non-uniformity 8,33,34 .Mix-and-compact method has been preferred over other considering their tendency of generating better contact between cement and treated sand.
The study contains 9 consolidated samples which are combinations of different sand particle sizes and concentration.From visual inspection and intensity of colors in Fig. 8, it can be observed that from lower to higher concentration, there is more precipitation in pore space.However, those pictures not only help to detect them but also to see how they are distributed along the samples.From fine to coarse sand, spatial distribution of carbonate minerals differs.Especially, for fine sand, uniform distribution of precipitates is observed.Another essential fact is that in spite of the same curing and treatment methodology, coarse sand has cracked from the higher half part.It can be explained by non-uniform consolidation and nature of big sand particles which requires bigger precipitates to bind them together.
Another factor to address is how amount and degree of uniformity of precipitation affects the strength of samples.Figure 9 generalizes this phenomenon for 3 different sand types.Minimum and maximum ISE strength are reflected on the figure to highlight the importance of strength distribution.Generally, it can be observed that the difference can be up to 4,000 psi for coarse sand and this value can rocket to 8,500 psi if particle size is around   www.nature.com/scientificreports/250 and 475 μm.On the other hand, fine sand is showing more stable consolidation with a maximum of 3,000 psi for high concentration and less than 1,000 psi when concentration is further reduced.In comparison to other studies, ISE strength values are higher due to absence of fines which has negative effect on consolidation 35,36 .To emphasize concentration and particle size, Figs. 10 and 11 highlight the full distribution of ISE strength of samples.The former illustrates fine sand sample and its magnitude of strength with respect to concentration.In spite of difference in the yield and efficiency of solutions, all give rise to a strength nearly equal at top half of sample and proportional values to concentration at bottom half.The main point here is consolidation uniformity and consolidation potential of sample containing 125 and 250 μm sand particles.It is the range of value which yields uniform consolidation thanks to clogged connected porosity but limited strength up to 4,000 psi.
Figure 11 analyzes consolidation from the aspect of particle size.Highest concentration solutions are analyzed to clearly see the distribution of precipitation.It stands out from graph that when particle size becomes higher than 250 μm, probability of non-uniform distribution rise significantly.Maximum strength has jumped from 4,170 to nearly 9,000 psi.However, such high value of strength cannot be considered efficient because at least 10 times lower value at the top half of sample.Such lower values are also experienced in the top half of coarse sample, but they have twofold less strength at bottom.The difference between coarse and medium sand size can be explained by the limited contact area of coarse sample which has particle diameter with at least 425 μm.
One of the alternatives to address non-uniformity problem is to modify EICP content.Since most of proposed EICP solutions are water-based and drained off as they are injected to the area of interest.Therefore, additives such as biopolymers are introduced to improve viscosity and fluid retention abilities without harming efficiency of reaction.Furthermore, it is important to note that even though the proposed EICP solution provided very high strength consolidated samples, for it be an effective solution to the industry, it has to provide flowability (high permeability) to allow for the oil or gas to flow through this consolidated sand and therefore not harm the well's productivity.This aspect has not been studied in this work but is considered for future work.

Conclusion
From the discussion of temperature sensitivity and consolidation, the following conclusions can be drawn: • For highly saturated EICP solutions, higher temperature environment (up to 90 °C) can favor reaction.It is shown by ICP analysis that solutions which contain higher than 50,000 ppm of Ca 2+ , Mg 2+ metal ions, and 110,000 urea are especially characterized by sensitiveness to high temperature environments.• In terms of efficiency, 35,000 ppm of metal ions is threshold value to achieve nearly complete 80% conver- sion to precipitation.Higher than this value the conversion is limited to 50% but favored with controllability which can positively affect application of EICP solution.• Despite numerous studies pinpointing Mg 2+ positive effect on precipitation, for highly saturated solution, they performed less reactive nature.However, their presence contributes to generation of aragonite which can be debatable in terms applicability in consolidation process.• From visual inspection of scratched sand sample, most of the treated samples exhibit non-uniform distri- bution of precipitation.Fine sand of particle size between 125 and 250 µm exhibits comparatively more uniformity.• The relationship between mass of precipitation and ISE strength is strongly non-linear and function of particle size.Medium sand size experiences 2 times increase in ISE strength with an increment of 1.1% of cumulative Ca 2+ and Mg 2+ concentration.

Figure 8 .
Figure 8. Visual inspection of treated sand samples.

Figure 9 .
Figure 9. Basic statistical analysis of strength.

Figure 10 .
Figure 10.Longitudinal strength distribution of high concentration samples.

Figure 11 .
Figure 11.Longitudinal strength distribution of fine sand.

Table 1 .
27CP solution materials and their functions.whichisproportional to concentration27.Since the equipment can detect up to 100 ppm concentration, it has been calibrated with 4 specific calibration fluids of 1, 10, 25 and 50 ppm.

Table 2 .
Reactants.EICP solutions which contain 1 to 7 volume ratios of reactant to catalyzer an overall constant volume of 40 ml.Lastly, mix them thoroughly.