Read without ads and support Scribd by becoming a Scribd Premium Reader.
See Premium Plans
 
NMR and Viscosity Investigation of Clathrate Hydrate Formationand Dissociation
Shuqiang Gao,
†
Walter G. Chapman,*
,†
and Waylon House*
,‡
Chemical Engineering Department, Rice University, Houston, Texas 77251, and Petroleum Engineering Department, Texas Tech University, Lubbock, Texas 79406
To better understand clathrate hydrate mechanisms, nuclear magnetic resonance (NMR) and viscosity measurements were employed to investigate tetrahydrofuran (THF) hydrate formationand dissociation processes. In NMR experiments, the proton spin lattice relaxation time (
T 
1
) of THF in deuterium oxide (D
2
O) was measured as the sample was cooled from room temperaturedown to the hydrate formation region. The D
2
O structural change around THF during this processwas examined by monitoring the rotational activation energy of THF, which can be obtainedfrom the slope of ln(1/ 
T 
1
) vs 1/ 
T 
. No evidence of hydrate precursor formation in the hydrateregion was found.
T 
1
measurements of THF under constant subcooling temperature indicatethat THF hydration shells do not undergo much structural rearrangement during induction.The
T 
1
of THF was also measured as the sample was warmed back to room temperature afterhydrate dissociation.
T 
1
values of THF after hydrate dissociation were consistently smaller thanthose before hydrate formation and never returned to original values. It was proposed that thisdifference in
T 
1
after hydrate dissociation indicates that the THF
-
D
2
O solution is moremicroscopically homogeneous than before hydrate formation. In viscosity experiments, aChampion Technologies hydrate rocking cell (CTHRC) was used to probe the residual viscosityphenomenon after Green Canyon (GC) gas hydrate as well as THF hydrate dissociation. Theresidual viscosity reported in the literature was observed after GC hydrate dissociation but notafter THF hydrate dissociation. Because GC hydrate behavior involves significant amounts of gas mass transfer while THF hydrate does not, one might conclude that the residual viscosityobserved after GC hydrate dissociation was likely caused by the supersaturated gas concentrationand its general effect on solvent viscosity, not necessarily by a clathrate water structure lingeringfrom the solid.
1. Introduction
Gas hydrates, also known as clathrate hydrates, arenonstoichiometric crystalline compounds composed of cooperative hydrogen-bonded water molecules formingnanoscale cage-like structures, which accommodatesmaller guest molecules in nearly spherical cavities.Depending on the size and the composition of the guestmolecules, gas hydrates usually form in at least threetypes of crystal structures, I, II, and H.
1
It has beenwidely recognized that gas hydrates are significantfactors in flow assurance,
2
global warming,
3
and seafloorstability
4
and are a potential energy supply.
5
Gashydrates have also been proposed for use in gas separa-tion,
6
transportation, and storage.
7
Since the first discovery that water forms clathratehydrate with chloride in 1810 by Sir Humphrey Davy,about 200 years of research has been devoted tounderstanding hydrate phenomenon. This continuouseffort has well-established the thermodynamic, physical,and structural properties of gas hydrates and discovereda rich collection of hydrate formers. However, clathratehydrate formation and decomposition mechanisms andkinetics are still far from clear due to a lack of appropri-ate experimental techniques capable of probing dynamicstructural information on the molecular level. Theactual formation and dissociation mechanisms haveimportant impacts on all hydrate applications. Clarifi-cation of the formation mechanism is especially impor-tant for designing kinetic inhibitors or antiagglomeranthydrate inhibitors that are expected to intervene withthe hydrate formation process on the molecular scale.Current hypotheses
1
about hydrate formation anddissociation involve hydrate precursor and postcursorstructures. The hypothesis states that, prior to hydrateformation, water molecules form individual clathratecages (hydrate precursors) around the dissolved guestmolecules with one guest molecule inside each cage.Under favorable conditions, these cages will agglomer-ate and form solid clathrate hydrate. The hypothesisalso suggests that, upon melting, the solid hydratephase dissolves into hydrate cage clusters. Those clus-ters, also known as residual structures or postcursors,persist in the liquid phase over a long period of timeprovided the temperature stays within certain limits.
8
To understand hydrate formation and dissociationmechanisms, many researchers have studied hydrateswith a particular emphasis on the precursor and post-cursor hypotheses. Molecular simulations generated voluminous numerical data on the elemental moleculardetail of hydrophobic hydration processes, which isimportant for evaluating the precursor hypothesis. A vigorous debate continues on whether hydrophobicsolutes structure the water molecules around them. One
* To whom correspondence should be addressed. W.G.C:phone, (713) 348-4900; fax, (713) 348-5478; e-mail, wgchap@rice.edu. W.H.: phone, (806) 742-3573; fax, (806) 742-3502;e-mail, Waylon.House@ttu.edu.
†
Rice University.
‡
Texas Tech University.
7373
 Ind. Eng. Chem. Res.
2005,
44,
7373
-
7379
10.1021/ie050464b CCC: $30.25 © 2005 American Chemical SocietyPublished on Web 08/09/2005
 
 view argues that nonpolar solutes enhance the order of the hydrogen-bonding network and create structuredhydration shells around them.
9
-
20
On the other hand,some simulation results showed no indication of en-hanced structure around hydrophobic solutes.
21,22
Eventhough computer simulation studies have reached con-flicting conclusions on whether nonpolar solutes struc-ture water network around them, they all seem to agreethat long-lived clathrate cages in solution do not exist.The orientation of hydrogen bonds in the hydration shellis qualitatively the same as in hydrate clathrate cages,but the organization of the hydrogen bonds is lessordered and the number of water molecules involved ineach hydration shell is also different from that in aclathrate hydrate cage. Experimental work has alsobeen carried out to probe the water structures arounddissolved nonpolar solutes. Data seem to agree withthose from computer simulation; that is, only looselyorganized hydration cages form around hydrophobicsolutes, with no evidence of regular clathrate cages insolution.
23
-
27
The fact that no regular clathrate cages have beenexperimentally confirmed in the water phase beforehydrate formation indicates that the actual hydrateformation process may be much more complicated thanthe hypothesis suggests. Studies of molecular hydrateformation mechanisms present great experimental chal-lenges due to the low equilibrium concentrations of gases in water and the high pressures required. Experi-ments usually require specially designed sample cellsand high sensitivity instruments capable of providingstructural information on a molecular scale.NMR
T 
1
measurement has been shown to be apowerful technique in studying microdynamic behaviorof liquids
28
and providing local molecular-level struc-tural information surrounding the NMR respondingguest molecules. THF and water form structure IIhydrate at
∼
4.5 °C under ambient pressure. It is a goodmodel system to study clathrate hydrate mechanismsbecause its formation conditions are mild and THF ismiscible with water at the conditions of interest. In thiswork, proton
T 
1
of THF in D
2
O was measured as thesample was cooled from room temperature to hydrateformation conditions. This allowed us to monitor thestructural change of the hydration shells around THFmolecules as a function of temperature and provided anopportunity to examine whether hydrate precursorsform under hydrate conditions.The residual structure dissociation hypothesis is notas widely studied as the hydrate precursor hypothesis.It has been observed that re-formation of gas hydratesfrom this “hydrate melt” requires less subcooling andless induction time. This phenomenon is called the
memory effect
. Vysniauskas and Bishnoi
29
proposed thatit is the existence of residual clathrate structure inwater after hydrate dissociation that is responsible forthe memory effect.More recently, Sloan et al.
30
discovered that the viscosity of water after methane hydrate dissociationis initially higher than that before hydrate formationunder the same pressure and temperature. The viscosityreturns to normal only after a long period of time. Thismacroscopic residual viscosity phenomenon is arguedas evidence of residual structure, i.e., clathrate ag-gregates remaining in water after gas molecules arereleased.
31
However, it is well-known that the diffusionrate of methane in water is very small. After hydratedissociation, it is very likely there is still a large amountof methane dissolved in water solution due to masstransfer rate limitations. Since dissolved methane struc-tures the hydrogen-bonding network in water,
32
watersupersaturated with methane will have a higher viscos-ity than water simply saturated with methane underthe same pressure and temperature. Therefore, it is notclear whether the observed residual viscosity is causedby remnant clathrate structures or an excess amountof dissolved methane.To eliminate the complication of mass transfer phe-nomenon on the experimental results mentioned aboveand obtain a better understanding of the residualhydrate structure hypothesis, in this work we chose tostudy the possibility of residual viscosity after THFhydrate dissociation. THF is miscible with water at theconcentration of solid THF hydrate;
33,34
thus, experi-ments can be performed at constant concentrationbefore, during, and after hydrate formation. Stangretand Gampe
35
recently demonstrated that the hydropho-bic hydration of THF definitely dominates the hydro-philic one at low THF concentrations. This is the caseat hydrate composition THF:H
2
O
)
1:17 (molar), sowater
-
THF hydrogen bonds are not very important forthe fluid structure of dilute solutions. Therefore, interms of water structure change, THF is not far fromother nonpolar solutes. Viscosity of the sample was monitored before hydrateformation and after dissociation using a ChampionTechnologies hydrate rocking cell (CTHRC) (Figure 1)to detect the reported residual viscosity phenomenon.
30
Since THF concentration in water is constant, if theresidual viscosity phenomenon is observed, it would bea strong demonstration of the existence of residualclathrate hydrate structures, i.e., postcursors.Fleyfel et al.
36
combined macroscopic hydrate experi-ments (visual rocking cell) with microscopic hydrateexperiments (NMR) to investigate clathrate residualstructure at the point where hydrate particles becomeinvisible. They found that some clathrate cages mightstill exist in solution after hydrates visually disappear.In this work, to investigate the molecular evidence of residual structure for THF hydrate, we measured
T 
1
 values of THF as a function of temperature after THFhydrate dissociation in the NMR experiment mentionedearlier. Results were compared with that before hydrate
Figure 1.
Schematic of the Champion Technologies hydraterocking cell apparatus.
7374
Ind. Eng. Chem. Res., Vol. 44, No. 19, 2005
 
formation to detect any difference caused by the historyof hydrate formation.
2. Experimental Section2.1.NMRMicroscopicMeasurements.2.1.1.Theo-retical Background.
Spin lattice relaxation mecha-nisms for THF are composed of two parts, intermolecu-lar dipole
-
dipole and intramolecular dipole
-
dipoleinteractions. When the THF to D
2
O molar ratio is 1:17(the same composition as in THF hydrate), most of theTHF molecules are completely surrounded by D
2
Ohydration shells, and intra-dipole
-
dipole is the mainspin relaxation mechanism. Orientational structures of D
2
Os in hydration shells have direct impact on therotational motion of enclosed THF molecules. On thebasis of NMR theory,
37
T 
1
is inversely proportional tothe rotational correlation time
τ 
c
.
τ 
c
is usually believedto follow the Arrhenius behaviorwhere
τ 
o
is a constant preexponential factor;
E
a
is therotational activation energy, a manifestation of hydra-tion cage structure around THF and not sensitive totemperature;
T 
is the temperature in kelvin; and
R
isthe molar gas constant. Combining eqs 1 and 2 yieldseq 3.The slope of ln(1/ 
T 
1
) versus (1/ 
T 
) gives the rotationalactivation energy
E
a
. Therefore, by tracking the changesin the slope of ln(1/ 
T 
1
) versus (1/ 
T 
), we can measure
E
a
and evaluate the changes in hydration shell structurearound dissolved THF molecules as a function of tem-perature.
2.1.2. Experimental Details.
T 
1
measurements of THF (Aldrich, 99
+
%) in D
2
O (Cambridge Isotope Labo-ratories, D 99.9%) at various temperatures were per-formed on an 85-MHz Oxford horizontal 31 cm wide boreNMR with imaging capability, using a LITZ RF VolumeCoil (with 14 cm internal diameter) from Doty Scientific,Inc. (Figure 2). Data were acquired and processed using Varian VNMR software and INOVA hardware systems. An Air-Jet temperature controller blew dry and cold airto control the sample temperature. It is capable of controlling temperature from
-
40 to 100 °C with
(
0.1°C stability. A glass bottle with a cap that has a Teflonliner was used to contain the deoxygenated THF:D
2
O
)
1:17 (molar ratio) mixture. A LUXTRON fluoropticthermometer was mounted into the glass containerthrough the cap to monitor system temperature. Itsoutput reading resolution is 0.1 °C.Since trace amounts of oxygen may alter the
T 
1
of THF significantly, we deoxygenated pure D
2
O and THFliquids separately in Teflon containers by periodicallyflushing the gas above the liquid phase with purenitrogen gas while periodically shaking the samplecontainers to accelerate the diffusion of oxygen from theliquid samples. After flushing six or seven times overabout 12 h, THF and D
2
O were mixed on the molar basisof 1:17. All these processes were operated within aclosed glovebox with nitrogen environment. After al-lowing the sample to stabilize for half a day, we put thesample into the NMR probe for
T 
1
measurements. (Thesuccess of deoxygenation was demonstrated by that factthat, after deoxygenation,
T 
1
increased about 2 scompared to the sample without deoxygenation).
T 
1
 values were measured using the inversion recoverytechnique. All samples’ 90° and 180° pulses werecalibrated before every measurement. VNMR software,given inputs of possible minimum and maximum
T 
1
 values, automatically generates standard 180°
-
τ 
-
90°pulse sequences with various values of 
τ 
. It took 4
-
6min to take a
T 
1
data point.The sample was cooled from room temperature (
∼
25°C) to subzero temperatures with an average coolingrate of 
∼
0.5 C°/h until hydrate formation occurred.Hydrate formation was indicated by a jump in thesample temperature. Oyama et al.
38
proposed that,during the induction period, water around dissolved CO
2
takes time to rearrange into individual clathrate cagesbefore hydrate formation. To test this hypothesis, wemeasured
T 
1
as a function of time while the temperaturewas kept constant in subcooling state. If the hydrationshell structures around THF experience rearrange-ments, the
T 
1
of THF would vary. After hydrate forma-tion, temperature was raised to completely dissociatethe hydrate.
T 
1
was also measured as the sample washeated back to about 25 °C.
2.2. Viscosity Macroscopic Measurements.
The viscosity experiments for the THF (99
+
%, Aldrich)
-
deionized (DI) water system were performed on a state-of-the-art CTHRC, which is an apparatus with high-pressure sapphire tubes placed on a rack. It is capableof handling pressures up to 5000 psi. The rack isimmersed in a temperature-controlled thermal bath andit rocks
(
45° from horizontal position through a com-puter-controlled step motor. As the rack rocks, a ballinside the tube rolls from one end to the other, and eachtube functions such as a rolling ball viscometer. Themotion of the ball is detected through two sensorsmounted on both ends of each tube. The time that theball takes to travel from one sensor to the other isrecorded as an indication of the fluid viscosity insidethe tube. Changes of the fluid viscosity in the tubes canbe sensitively detected by measuring the ball travelspeeds. Hydrate formation is indicated by an increasein fluid apparent viscosity while system temperatureand pressure were kept constant. The ball was eventu-ally immobilized by the hydrate blockage. Hydrate onsetcan also be detected by visual observation of hydratecrystal deposit on thetube’s inside wall.Balltraveltime,temperature, and pressure data are collected everyminute into a terminal computer using LabView.
Figure 2.
Schematic of the NMR experimental setup.
1
T 
1
∝
τ 
c
(1)
τ 
c
)
τ 
o
e
 E
a
 / 
 RT 
(2)d ln(1/ 
T 
1
)d(1/ 
T 
)
)
 E
a
 R
(3)
Ind. Eng. Chem. Res., Vol. 44, No. 19, 2005
7375
Search
Search History:
Searching...
Result 00 of 00
00 results for result for
  • p.
    • Notes
    Load more