FIRST INVESTIGATION OF MODIFIED POLY(2-VINYL-4,4-DIMETHYLAZLACTONE)S AS KINETIC HYDRATE INHIBITORS
Authors: Lilian H. S. Ree,aMalcolm A. Kelland,aPeter J. Rothb,c and Rhiannon Batchelorc
aDepartment of Mathematics and Natural Science, Faculty of Science and Technology, University of Stavanger, N-4036 Stavanger, Norway
b Department of Chemistry, Faculty of Engineering and Physical Sciences, University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom
c Centre for Advanced Macromolecular Design, University of New South Wales, Sydney, Australia
Key words:petroleum;low dosage hydrate inhibitor; kinetic hydrate inhibitor;hydrate; polymers; kinetics.
ABSTRACT
A series of polymers of 2-vinyl-4,4-dimethylazlactone (VDMA) of varying molecular weights modified with small organic amines have been synthesized. Their performance as kinetic hydrate inhibitors (KHIs) has been investigated in high pressure steel rocking cells using a Structure II-forming synthetic hydrocarbon gas mixture. It was found that the PVDMA polymer with the lowest molecular weight (Mn = 1845 g/mole by 1H NMR) performed the best. It was also found that the n-propylamine derivative performed the best of the amine derivatives. The cloud point of this polymer derivative was found to be lower than ambient temperature, which is considered too low for practical oilfield applications. However, high cloud point PVDMA derivatives such as the ethylamine or pyrrolidine derivatives still gave reasonable KHI performance. The KHI performance of the n-propylamine derivative of PVDMA-I was tested at different concentrations in the range 1000-7000 ppm. It was found that the performance improved as the polymer concentration was increased.
1INTRODUCTION
Gas hydrate pluggingis a major flow assurance issue to consider in the oil and gas industry, particularly for flow lines in subsea and cold climate situations. (Carroll, 2009; Makogon, 1997; Sloan and Koh, 2008) One of the methods to combat this problem is the use of kinetic hydrate inhibitors (KHIs). (Kelland, 2006; Kelland, 2011; Perrin et al, 2013). KHIs are one class of low dosage hydrate inhibitors (LDHIs) which, as the name implies, are added at low concentrations in the order of 0.5-2 wt.% as formulations in solvents. KHIs are often a mixture of chemicals but always with one or more water-soluble polymers as the main active ingredients. Synergists may be added, which may include other polymers or non-polymeric chemicals including the solvent(s) used in the formulation. Many polymers have been investigated for their efficacy as KHIs but only a few have been commercialized. This includes polymers based on the monomers N-vinyl pyrrolidone, N-vinyl caprolactam (VCap) and N-isopropylmethacrylamide (NIPMAM) as well as hyperbranched poly(ester amide)s. One of the field compatibility issues with KHI polymers is that they often have limited solubility in produced water especially at high temperatures and salinities. (Kelland, 2014) Above a certain temperature called the cloud point (TCl) the polymer will often phase separate, giving a cloudy solution. Further heating to usually an incrementally higher temperature (TDp) can lead to precipitation of the polymer, which may then deposit and build up in the flow line near the injector site. Operator and service companies would consider this detrimental as the polymer is then apparently unavailable in the water phase for carrying out kinetic hydrate inhibition. Although field cases exist where a KHI polymer has been injected into well stream temperatures above the cloud and deposition point, and successfully used to prevent hydrate formation, most operators prefer to avoid this complication by using a product that will remain water-soluble or at worst water-dispersible at the well head temperature.
Poly(N-vinyl caprolactam) (PVCap) and poly(N-isopropylmethacrylamide) (PNIPMAM) have cloud points around 30-45oC depending on the method of manufacture. This is often too low for injection at the well head temperature. One way to overcome this problem is to add chemicals that often also act as synergists, to raise the deposition point of the KHI formulation when injected.(Kelland, 2014)Another method is to copolymerize the active monomers with a more hydrophilic comonomer. For example, VCap or NIPMAM copolymerize in a 1:1 molar ratio with N-methyl-N-vinyl acetamide (VIMA) with optimized molecular weight distributions have been shown to give similar or superior performance KHIs to the homopolymers but with significantly higher cloud and deposition point temperatures. (Kolle et al., 1999; Talley and Mitchell, 1998)Another method is to build into the active monomer a more hydrophilic functional group so that the new homopolymer will be water-soluble at higher temperatures than the unmodified polymer. An example of this is the use of dimethylhydrazido groups instead of isopropylamido groups in substituted acrylamide monomers (Figure 1).(Mady and Kelland, 2014) The extra nitrogen atom in the hydrazido group confers a greater hydrophilicity on the monomer than just one nitrogen atom from an amide group. Thus, poly(dimethylhydrazidoacrylamide) has no cloud point in water as a 1 wt.% aqueous solution at any temperature up to 100oC. This homopolymer or copolymers with NIPMAM or VCap are still very active KHIs with improved cloud points compared to PVCap or PNIPMAM. (Mady and Kelland, 2015).
Figure 1. Structures of polyIPAM, polyIPMAM and the hydrazine-based analogues poly(N,N-dimethyhydrazidoacrylamide derivatives R= H or CH3 .
A related example of modifying a monomer to raise the cloud point of its polymers is the alkylated polyacrylamidopropylsulfonates(alkylated polyAMPS) (Figure 2). (Peiffer et al, 1999)A hydrophobic pendant alkyl group alone would not give a water-soluble polymer but incorporation of a hydrophilic sulfonate group made this polymer class water-soluble as long as the alkyl chains R1 and R2 were not too large. Conversely, the AMPS homopolymer alone is a poor KHI, but it was found that the addition of a hydrophobic tail between the amide and the sulfonic acid group increased the performance of the polymer. The optimum size of the tail R1 was 5 carbon atoms long. A homopolymer of this monomer gave a KHI performance similar to PVCap in a high pressure natural gasmini-loop.
Figure 2. Alkylated AMPs polymers, where R1 is an alkyl tail of 1-6 carbon atoms and R2is H or CH3.
We were initially interested in poly(2-vinyl-4,4-dimethylazlactone) (PVDMA) as the azlactone pendant ring resembled the oxazoline ring that had been investigated earlier as a functional groups for KHI polymers in polyvinyloxazolines (Figure 3). (Reyes et al.; 2013; Colle et al., 1996)The monomer 2-vinyl-4,4-dimethylazlactone is commercially available but not on an industrial scale.
Figure 3. Poly(2-vinyl-4,4-dimethylazlactone) (PVDMA) (left) andring-closed polyvinyloxazolines, R = H or CH3 (right).
However, PVDMA is not water-soluble as the homopolymer, although it was known that PVDMA hydrolyses to give poly(N-acryloyl-2-methylalanine) (PAMA) with both pendant dimethyl groups and a carboxylate group (Figure 4) (Gardner et al, 2012). The resemblance of PAMA to the alkylated AMPS polymers described earlier is apparent. Therefore, we were interested in investigating PAMA as a KHI. Since the size of the 2,2-dimethyl hydrophobic group was relatively small we expected the KHI performance to be low, and so we sought ways to incorporate larger hydrophobic groups into the polymer. One method is to begin with a 2-vinyl-4,4-dialkylazlactone monomer where the alkyl groups are larger than methyl. However, so far we have not explored this route. Instead, we chose a route explored already by some of us by incorporating hydrophobic pendant groups through postpolymerization modification of PVDMA with primary or secondary amines. (Figure 5)This ring-opening addition reaction proceeds quantitatively under mild conditions and produces acrylamide polymers with pendant alkyl or dialkylamide groups. (Heilman et al., 2001; Ho et al., 2012). Compared to the related poly(N-alkylacrylamide)s and poly(N,N-dialkylacrylamide)s, the PVDMA-derived species feature a bisamide structurewith the introduced alkyl groups further away from the polyvinyl backbone and therefore with greater degrees of rotational and vibrational freedom (entropy).As the postpolymerization modification of PVDMA occurs at the pendant groups, all daughter polymers derived from one reactive precursor feature identical degrees of polymerization. This synthetic strategy is therefore ideal to study structure–property relationships of polymer series with the same degree of polymerization. Additionally, several PVDMA-derived bisamide species were recently shown to have tunable cloud points in aqueous solution. (Zhu et al., 2013; Quek et al., 2013; Pei et al.; 2015).As such, PVDMA presented a promising reactive platform to investigate the KHI performance of polyPAMA as well as a range of novel PVDMA-derived bisamidesof various degrees of polymerizationin high pressure rocking cells with a natural gas mixture.
Figure 4. Reaction scheme for hydrolysis of poly(2-vinyl-4,4-dimethylazlactone) (PVDMA) to the sodium salt of poly(N-acryloyl-2-methylalanine) (PAMA).
Figure 5. Reaction ofpoly(2-vinyl-4,4-dimethylazlactone) (PVDMA) with primary or secondary amines.
2EXPERIMENTAL
2.1Chemicals
All chemicals and solvents were purchased from VWR and Sigma-Aldrich and were used without further purification.
Four PVDMA precursors of different molecular weight were preparedby the RAFT processfollowing a literature procedure. (Moad et al., 2012; Zhu et al., 2013) Briefly, VDMA monomer, RAFT agent benzylpropyltrithiocarbonate, acetonitrile, and 4,4’-azobisisobutyronitrile as radical initiator were combined and heated to 70°C under inert atmosphere for 7 hours. PVDMA samples I–IV were prepared by varying the ratio of monomer to chain transfer agent (seeTable 1). Polymers were isolated by repeated precipitation into methanol.Monomer conversions ranged between 44–80%.Polymer molecular weights were calculated from monomer conversion determined by 1H NMR spectroscopy.Results are given in Table 1.
Table 1. Molecular weight of the provided polymers.
Polymer / Target Degree of Polymerization / Actual Degree of Polymerizationa / Number Average Theoretical Molecular Weight [g/mol]aPVDMA-I / 15 / 11 / 1845
PVDMA-II / 20 / 16 / 2450
PVDMA-III / 80 / 51 / 7350
PVDMA-IV / 200 / 87 / 12400
a calculated from monomer conversion determined by 1H NMR spectroscopy.
2.2Hydrolysis of PVDMA
PVDMA-II was hydrolyzed following a procedure known from the literature.1In a typical reaction PVDMA-II (97.5 mg, 0.70 mmol) was added to a round bottom flask. The polymer was dissolved in THF (3.5 ml), and aqueous NaOH (1 M, 0.35 mL, 0.5 eq) was added. The solution was stirred at room temperature for 15 minutes before additional aqueous NaOH (1 M, 0.35 mL, 0.5 eq) was added. The solution was stirred at room temperature for 30 minutes. Then the solvent was removed in vacuo.
2.3Modification of PVDMA with different amines
The VDMA polymers of different molecular weight were modified using different low molecular weight amines by following a procedure known from the literature.2 The example shown here is for making the isopropylamine derivative of PVDMA. PVDMA (0.150 g, 1.08 mmol of monomer units) was dissolved in DMF (4.3 ml) and acrylamide (0.008 g, 0.11 mmol) was added. DMF (2.3 ml) and isopropylamine (0.191 g, 3.23 mmol, 3.0 eq) were mixed in a separate vial, and thenadded to the polymer solution. The reaction mixture was stirred at room temperature overnight. The polymer was precipitated in diethyl ether twice, and was then dried in vacuo.The different modified polymers were isolated in high yields (75-98%).Figure 6 summarizes all the amide derivatives of PVDMA that were synthesized using this method.
2.4Determination of Cloud Point (TCl)
A 2500 ppm solution of polymer in deionized water was carefully heated at about 2 °C/min while making visual observations throughout. The temperature at which the first sign of haze in the solution was observed was determined as the cloud point (TCl). (Kjøniksen, et al.; 2005). The test was repeated for reproducibility.
Figure 6. Amide derivatives of PVDMA synthesized. Top row, left to right: methyl, ethyl, iso-propyl, n.-propyl. Bottom row, left to right: sec-butyl, dimethyl, diethyl and pyrroldinyl.
2.5High Pressure KHI Experimental Methods
Performance testing of the modified VDMA polymers as KHIs was conducted in a high pressure gas hydrate rocker rig experiment as previously described by our research group.(Chua and Kelland, 2011). The equipment was manufactured and supplied by PSL Systemtechnik, Germany (7) and consists of five 40 mL high pressure steel rocking cells each containing a steel ball. The cells are positioned in a cooling bath during the experiments. A standard synthetic natural gas mixture (SNG) known to preferentially form Structure II gas hydrates was used in the experiments. The gas composition is given inTable 2.
Figure 7.The high pressure rocker rig with the five steel rocking cells positioned in the cooling bath.
Table 2.Composition of synthetic natural gas (SNG) used in the experiments.
Component / mole%Methane / 80.67
Ethane / 10.20
Propane / 4.90
iso-Butane / 1.53
n-Butane / 0.76
N2 / 0.10
CO2 / 1.84
The constant cooling KHI performance test procedure was as follows:
1. Each cell was filled with 20 mL of deionized water in which the sample had been dissolved to the desired concentration.
2. Air in the cells was removed by applying vacuum, then filling the cells with SNG to 3-5 bar. The cells were depressurized before a repeated step of vacuum was applied.
3. The cells were pressurized to approximately 76 bar with SNG.
4. The cells were rocked at 20 rocks per minute at an angle of 40° while they were cooled from 20.5 °C to 2.0 °C at a rate of 1.0 °C/h.
5. The pressure and temperature for each individual cell, and the temperature of the cooling bath were logged on a local computer.
Figure8 shows an example of the results from a KHI experiment done under constant cooling conditions, where pressure and temperature are plotted against time for each cell.A pressure drop of approximately 2 bar can be seen in the beginning of each experiment as gas dissolves in the aqueous phase. Because each cell is a closed system during the test, a linear pressure decrease is seen as the temperature is reduced. The first deviation from this linear pressure decrease is determined as the observed macroscopic onset temperature for hydrate formation, To. Nucleation might have occurred earlier on an undetectable scale. (Chen, et al. 2010; Qin et al. 2015).A rapid pressure decrease is seen at some time after the initial hydrate formation temperature, which indicates that hydrates are growing rapidly. The temperature at which the hydrate growth is at its most rapid, Ta, is determined at the steepest part of the graph. Figure 9 shows the pressure (P1) and temperature (T1) for cell 1, and how To and Tawas determined for the polymer solution in this cell. For each polymer 8-10 standard cooling tests were carried out. It was found that none of the cells gave any systematic errors that lead to consistently better or worse results than the other cells.As the To value refers to the first detection of hydrate formation, after which crystal growth can potentially lead to hydrate plugging, this is considered the most important of the two temperature parameters. The difference between the To and Tavalues gives some indication of the ability of the additive to retard the gas hydrate crystal growth processalthough the onset temperature has to be considered when comparing hydrate growth rates.
Figure 8. Example of pressure and temperature data versus time for five cells in a standard constant cooling KHI experiment. Cell 1 contained PVDMA-II modified with isopropyl amine, while cells 2-5 contained PVDMA-III modified with pyrrolidine.
Figure 9. Determination of To and Ta after a standard constant cooling experiment.
3RESULTS AND DISCUSSION
3.1KHI performance of hydrolyzedPVDMA
As PVDMA homopolymer is not soluble in water, the polymer was hydrolyzed as shown in Figure 4 prior to testing to make it water-soluble.The results from the KHI performance test of PAMA-II made fromhydrolyzing PVDMA-II are presented in Table 3 and compared to the results for deionized water (no polymer).
Table 3. Average onset (To) and rapid hydrate formation (Ta) temperatures for deionized water, and hydrolyzed PVDMA-II, PVCap, PNIPAM at 2500 ppm.
Polymer / Concentration [ppm] / To(av)[°C] / Ta(av)[°C] / To(av) – Ta(av)[°C]No polymer / - / 17.4 / 16.0 / 1.4
PAMA-II / 2500a / 17.0 / 15.8 / 1.2
PVCap 4k / 2500 / 10.6 / 9.1 / 1.5
PNIPAM 7k / 2500 / 10.9 / 7.5 / 3.4
aAverage of three results.
As can be seen from Table 3 the performance of PAMA-II at 2500 ppm was quite similar to the performance of pure, deionized water. This suggests that PAMA-II is not able to kinetically inhibit the formation of gas hydrates. In comparison we carried out experiments at the same conditions and the same equipment on a poly(N-vinyl caprolactam) (PVCap) (Mw = 4092 g/mole), as well as a low molecular weight poly(N-isopropylacrylamide) (PNIPAM) (Mw = 7353 g/mole) synthesized in our own laboratory.(Madyand Kelland, 2014; Mady and Kelland, 2015). Neither polymerization procedure or the final molecular weight distribution had been optimized for best performance but both gave reasonably low To values and thus reasonably good performances as KHIs compared to the negligible KHI effect of PAMA-II. Commercial KHI solutions that contain polymers such as PVCap give lower To values in our laboratory but we have chosen to compare PVCap and PNIPAM made in unoptimized processes in our laboratories as a better comparison to the KHI activity of the new PVDMA-based polymers which are also unoptimized. (Mady and Kelland, 2014; Reyes and Kelland, 2013).
3.2Performance comparison of the different modified PVDMAs
As PAMA-II failed to show any KHI performance we investigated modified versions of the polymer by reacting PVDMA of varying molecular weights with several low molecular weight amines in order to increase the hydrophobic portion of the polymers. After modification the KHI performance testing was carried out in high-pressure natural gas hydrate constant cooling rocking cells experiments. Table 4 summarizes the average onset (To) and fast hydrate formation temperatures (Ta) from 8-10 experiments. The experiments with no additive (i.e. pure deionized water, entry 1 in Table 3) are again included for comparison. The same data are presentedin Figure 10 to show the data scattering.Although there is a lot of data here, we thought it helpful to place all the results on one graph so any trends can be determined visually. In order to make Figure 10 easier to read all data corresponding to the same PVDMA has the same shaped marker. The markers with similar color refers to PVDMAs modified with the same amine. The diethylamine derivative of PVDMA-III (entry 16 in Table 4) was found to be insoluble in water, hence, the performance of this derivative was not tested.