WO2019010696A1

WO2019010696A1 - A method of gas-phase molecular layer-by-layer deposition on a microporous support

Info

Publication number: WO2019010696A1
Application number: PCT/CN2017/092938
Authority: WO
Inventors: Hua Wang; Daniel HIGGS; Steven George; Yanju Wang; David Moore
Original assignee: General Electric Company
Priority date: 2017-07-14
Filing date: 2017-07-14
Publication date: 2019-01-17

Abstract

The present disclosure provides methods of preparing a thin film composite membrane using gas-phase molecular-layer-by-layer deposition of reactive organic compounds, and thin film composite membranes. Depending on characteristics of the membrane and microporous support, the composite membrane may be used in RO, NF, FO, pervaporation, or gas separation processes.

Description

A METHOD OF GAS-PHASE MOLECULAR LAYER-BY-LAYER DEPOSITION ON A MICROPOROUS SUPPORT

GOVERNMENT LICENSE RIGHTS

This invention was made with government support under Contract DE-EE0005771 awarded by the U.S. Department of Energy. The government has certain rights in the invention.

FIELD

The present disclosure relates to methods of making a composite membrane using gas-phase molecular layer-by-layer deposition on a microporous support.

BACKGROUND

The following paragraphs are not an admission that anything discussed in them is prior art or part of the knowledge of persons skilled in the art.

Filtration membranes are used to separate fluid mixtures and solutions. For example, reverse osmosis (RO) , nanofiltration (NF) , and forward osmosis (FO) , may be used to remove dissolved ions, such as salts or minerals, from seawater, brackish water, dairy products, or oil and gas field brine. Gas separation membranes may be used to produce oxygen or nitrogen enriched air； purify natural gas by removal of nitrogen, hydrogen sulfide, or carbon dioxide； or separate carbon dioxide from power plant flue gas for sequestering.

Filtration membranes may include a polymeric thin film and a microporous support layer. The polymeric thin film provides the physical characteristics required to separate the desired components from the fluid mixture or solution, while the microporous support layer provides sufficient support to the polymeric thin film to operate under separation conditions. The filtration membrane may be formed by interfacial polymerization of monomers on the microporous support, or by molecular layer-by-layer deposition of reagents on the microporous support.

INTRODUCTION

The following introduction is intended to introduce the reader to this specification but not to define any invention. One or more inventions may reside in a combination or sub-combination of the apparatus elements or method steps described below or in other parts of this document. The inventors do not waive or disclaim their rights to any invention or inventions disclosed in this specification merely by not describing such other invention or inventions in the claims.

Interfacial polymerization forms the polymeric film on the microporous support layer by coating the support layer with an aqueous solution of a first reagent, then coating the support layer with a water-immiscible solution of a second reagent, and allowing the two reagents to polymerize at or near the fluid interface of the aqueous/water-immiscible solutions. Interfacial polymerization may generate non-uniform coatings, with film thickness of about 50 nm to about 5000 nm and surface roughness of about the same order of magnitude as the film thickness. The lack of precise control of the thin film thickness and roughness of interfacially polymerized membranes limit their performance since the flux of a thin film composite membrane is inversely proportional to the thin film thickness. Increased surface roughness contributes to increased fouling propensity.

Molecular layer-by-layer deposition (mLBL deposition, or MLD) forms the polymeric film by depositing nanometer scale coatings layer-by-layer through sequential, alternating reactions of reagents that react with the previously deposited reagents. The thickness of the film is controlled by the number of reaction cycles that are used. MLBL deposition on porous supports has been performed using solution-based synthetic reactions. Solution-based mLBL require coating and rinse solvents, such as toluene or acetone, which have adverse environmental impacts and/or require special disposal procedures. One example of a solution-based mLBL process is discussed in US20140083925A1. Vapor-based deposition of trimethyl aluminum and ozone has been used to grow Al₂O₃ atomic layer deposition (ALD) films on metallized polyethylene terephthalate (PET) substrates.

However, there remains the need for a vapor-based mLBL deposition method of organic monomer reagents on a porous support. Such a method may be used to form a membrane suitable for use in RO, NF, FO, pervaporation, or gas separation. For some supports, it would be desirable if such a vapor-based mLBL deposition method reduced or avoided the collapse and/or plugging of some or all pores in the porous support during the mLBL deposition process.

In one aspect, the present disclosure provides a gas-phase molecular layer-by-layer deposition method where organic reagents are deposited on a microporous support layer. Such a method includes repetitively depositing on to the microporous support, via a gas phase: (i) one or more first thin film composite membrane precursors, and (ii) one or more second thin film composite membrane precursors. The one or more first membrane precursors are reactive with the one or more second membrane precursors. In the method, the gas-phase deposition is performed isothermally at a temperature that is at or below the lowest thermal decomposition temperature of the first and second membrane precursors. The first and second membrane precursors have vapor pressures 0.1 Torr or higher at the temperature of the gas-phase deposition.

In another aspect, the present disclosure provides a thin film composite membrane. The membrane may be formed through the method discussed above. The membrane includes a microporous support having thereon molecular layers of (i) one or more first organic compound residues, and (ii) one or more second organic compound residues. Substantially all of the organic compound residues in each molecular layer are covalently bonded to one or more organic compound residues in one or more adjacent molecular layers. The membrane may have a crosslink density of at least 75％for RO thin film made of m-phenylene diamine and trimesoyl chloride precursors.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present disclosure will now be described, by way of example only, with reference to the attached Figures.

Figs. 1a and 1b show graphs illustrating the mLBL polyamide film thickness as a function of mPD and TMC flow coefficient on a non-porous support.

Fig. 2 shows a graph illustrating the mLBL polyamide film thickness as a function of drum speed on a non-porous support.

Fig. 3 shows a graph illustrating the polyamide film thickness as a function of the numbers of A/B reaction cycles on a non-porous support.

Fig. 4 shows a graph illustrating the IR spectrum of the mLBL polyamide thin film on a non-porous support.

Fig. 5 shows a graph illustrating the root-mean square (RMS) roughness as a function of cycle number.

Fig 6 shows a graph illustrating the ratio of roughness: thickness as a function of number of mLBL cycles.

DETAILED DESCRIPTION

Generally, the present disclosure provides methods of preparing a thin film composite membrane using gas-phase molecular-layer-by-layer deposition of reactive organic compounds, and thin film composite membranes. Depending on characteristics of the membrane and microporous support, the composite membrane may be used in RO, NF, FO,or gas separation processes.

In one example, amethod according to the present disclosure includes repetitively depositing on to the microporous support, via a gas phase: (i) one or more first thin film composite membrane precursors, and (ii) one or more second thin film composite membrane precursors

Although the present disclosure refers to “first” and “second” precursors， “first” and “second” are merely in relation to each other and it is not required that the “ first” precursor be deposited on the microporous support before anything that could be considered the “second” precursor. The “first” membrane precursor is deposited on a substrate. The “first” membrane precursor may covalent bond with a chemical functionality on the substrate， or may interact with the substrate without an apparent bond. The substrate may be: the microporous support (in which case the first precursor is deposited directly on the microporous support) ； or a molecular layer previously deposited on the microporous support (such as a molecular layer formed by the deposition of a compound that could be used as a second thin film composite membrane precursor) . In some other examples, the substrate may be an intermediate layer between the support and the first deposited molecular layer(regardless of whether the first deposited layer is composed of “first” or “second” precursors) .

The expression “repetitively depositing (i) one or more first membrane precursors and (ii) one or more second membrane precursors” should be understood to mean that:

i. a first membrane precursor, or a mixture of different first membrane precursors, is deposited on the substrate；

ii. a second membrane precursor, or a mixture of different second membrane precursors is then deposited on the resulting substrate；

iii. a first membrane precursor, or a mixture of different first membrane precursors is then deposited on the resulting substrate；

iv. a second membrane precursor, or a mixture of different second membrane precursors is then deposited on the resulting substrate；

v. etc.

The compound or compounds used in the deposition of one or more first membrane precursors (e.g. step iii, above) need not be identical to the compound or compounds used in previous depositions of one or more first membrane precursors (e.g. step i, above) . Similarly, the compound or compounds used in the deposition of one or more second membrane precursors (e.g. step iv, above) need not be identical to the compound or compounds used in previous depositions of one or more first membrane precursors (e.g. step ii, above) . For example, the first membrane precursor deposited in step i, above, could be trimesoyl chloride (TMC) ； the second membrane precursor deposited in step ii, above, could be meta-phenylene diamine (mPD) ； the first membrane precursor deposited in step iii, above, could be a mixture of TMC and terephthaloyl chloride (TPC) ； and the second membrane precursor deposited in step ii, above, could be a mixture of mPD and piperazine. Subsequent depositions of first membrane precursors could use TMC alone, amixture of TMC and TPC, any other first membrane precursor, or any mixture of first membrane precursors. Similarly, subsequent depositions of second membrane precursors could use mPD alone, amixture of mPD and piperazine, any other second membrane precursor, or any other mixture of second membrane precursors.

The one or more first membrane precursors are reactive with the one or more second membrane precursors. Since the precursors being deposited react with previously deposited precursors, it would be understood that the previously deposited precursors are the components making up the current reactive substrate since they previously reacted with components making up the previous reactive substrate. These repetitive depositions result in the formation of molecular layers that are covalent bonded to the previously deposited molecular layers. The present disclosure may equivalently refer to a “membrane precursor” ， a “precursor” ， an “organic compound” ， or a “reactive compound” . The deposited molecular layer may be referred to as the “substrate” ， or the “reactive substrate” ， and the chemicals making up the molecular layer may be referred to as the “compound residues” or “residues” .

For example, if a previously deposited molecular layer includes amine groups, the current substrate would include amine groups and the membrane precursor being deposited would include an amine-reactive functional group. If the membrane precursor included three amine-reactive functional groups, one or two of those reactive functional groups may form a bond with amine groups in the substrate, leaving one or two amine-reactive functional groups available for a subsequent reaction as a part of the reactive substrate.

In order to be able to transport the first and second membrane precursors via gas-phase without significant decomposition of the precursors, the gas-phase deposition is performed isothermally at a temperature that is at or below the lowest thermal decomposition temperature of the first and second membrane precursors. The deposition temperature is also selected to avoid degradation of the microporous support.

The temperature and pressure of the gas-phase deposition are selected to result in vapor pressures of the first and second membrane precursors being 0.1 Torr or higher.

When the vapor pressures are 1 Torr or higher, standard delivery methods can be used. For example: exposure can be controlled by timing the actuation of a fast-switching valve mounted between a source vessel and a delivery line. The self-termination can be observed by monitoring the deposition rate as a function of exposure time. Aneedle valve can also be added to regulate the flow and concentration of the precursor in the reactor.

When the vapor pressure is from 0.1 Torr to 1 Torr, an alternative delivery method is needed. For example: an inert gas can be bubbled into the source vessel in order to increase the surface exchange between liquid and gas and facilitate the reactant evaporation.

The gas-phase deposition operating conditions depend on the vapor pressure of the precursors being used in the method. Vapor pressure of a compound refers to the pressure exerted by a vapor in equilibrium with its condensed phases at a given temperature. At a given temperature, aprecursor with a lower vapor pressure may require a lower operating pressure than a precursor with a higher vapor pressure. For some precursors, the gas-phase deposition may be performed at atmospheric pressure. For other precursors, the gas-phase deposition may be performed at a pressure from about 0.1 Torr to about 10 Torr. Since vapor pressure is determined at a given temperature, the vapor pressures of the precursors may be raised by preforming the gas-phase deposition at an elevated temperature. Selecting a higher operating temperature may reclassify a “low vapor pressure precursor” as a “high vapor pressure precursor” ， which may allow the deposition to be operated at atmospheric pressure. In contrast, selecting a lower operating pressure may allow the deposition to be operated at a lower temperature, which may allow the use of an otherwise unavailable precursor (for example because the precursor degrades at the higher temperature) . The delivery method of a precursor may also be used to affect whether the deposition is performed at atmospheric operation. For example， a “pressure boost” approach for delivering low vapor pressure precursors may be used to control precursor dose, independent of reactor pressure. An example of such a delivery method is discussed by JS Jur and GN Parsons. ACS Appl. Mater. Interfaces, 2011, 3 (2) , 299-308.

Although the final operating conditions will depend on the precursors being used, the deposition may be performed: at a pressure of at least 0.01 Torr, such as from about 0.01 Torr to about 760 Torr (i.e. atmospheric pressure) , for example from about 0.05 Torr to about 100 Torr, or from about 0.1 to about 10 Torr； and at a temperature from about 30℃ to about 250℃, such as from about 60℃ to about 200℃, or from about 80℃ to about 150℃.

In some examples, the microporous support may be an amorphous polymeric support or a semi-crystalline polymeric support. Examples of amorphous polymeric supports include supports made with polybenzimidazole, polysulfone, and polyacrylonitrile. When the support is (a) an amorphous polymeric support and the operating temperature is close to the glass transition temperature of the support, or (b) asemi-crystalline polymeric support and the operating temperature is close to the melting temperature of the support, then the method may be performed with at least some of the pores of the support being at least partially full of a liquid that has, at the gas-phase deposition temperature, avapor pressure substantially below the gas-phase deposition pressure. For example, when the support is made from polysulfone (which has a Tg of about 180℃) or polyacrylonitrile (which has a Tg of about 95℃) , and the precursors are mPD and TMC (which may require an operating temperature from about 80℃ to about 150℃) , at least some of the pores of the support may be at least partially full of the liquid. When the support is made from polysulfone, and the operating temperature is greater than about 80℃, it is especially helpful to at least partially fill at least some of the pores of the support. It is even more helpful to at least partially fill at least some of the pores of the support when the operating temperature is greater than about 100℃, and even more helpful when the operating temperature is greater than about 120℃.

Performing the method with at least some of the pores being at least partially full of the liquid helps reduce or prevent pores from being blocked or collapsed at the operating temperatures and pressure. At least 50％of the pores of the microporous support may be at least partially full of the liquid. In some examples from about 80％to about 100％, and preferably about 100％, of the pores of the support are at least partially full of the liquid. Pores that are at least partially full of the liquid are preferably substantially completely full of the liquid. Microporous supports made from polymers having a higher glass transition temperature and a higher melting temperature than the operating temperature, or supports made from more structurally rigid material, such as ceramic-, zeolite-, or metal-microporous supports, do not need to be filled with the liquid in order to avoid pore blockage or collapse.

The liquid may be an ionic liquid that lacks functional groups reactive with the membrane precursors, for example 1-Butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl) imide. An ionic liquid should be understood to refer to an ionic compound that is a liquid below 100℃. Some ionic liquids are liquids below room temperature. An ionic liquid may have, as the ionic portion, a cation such as: 1) alkylammonium-, 2) dialkylimidazolium-, 3) phosphonium-or, 4) N-alkylpyridinium. Ionic liquids may have a low vapor pressure at room temperature, such as a vapor pressure of about 1 E-10 Pa (about 1 E-12 Torr) at 25℃. In some exemplary methods, an ionic liquid having a low solubility, or no solubility, for any of the membrane precursors is used. Using such an ionic liquid allows for a linear growth rate of the molecular layer-by-layer thin film.

The expression “microporous support” would be understood to refer to a support that has a plurality of pores where the average diameter is from about 1 nm to about 1μm. If the microporous support is an ultrafiltration membrane, the support may have a pore size from about 1 nm to about 0.05μm. If the microporous support is a microfiltration membrane, the support may have a pore size from about 0.05μm to about 1μm. If the composite membrane is intended for use in nanofiltration or reverse osmosis, the microporous support may have a pore size from about 5 nm to about 50 nm.

The microporous support may be a glassy microporous polymeric support having a Tg at least 20℃ higher, such as at least 60℃ or 80℃ higher, than the gas-phase deposition temperature. The microporous support may be a semi-crystalline microporous polymeric support having a Tm at least 20℃ higher, such as at least 60℃ or 80℃ higher, than the gas-phase deposition temperature. In some examples, the microporous support may be: a polysulfone microporous support, a polyethersulfone microporous support, a polyvinylidene fluoride microporous support, a nylon microporous support, a polycarbonate microporous support, a polyacrylonitrile microporous support, a polytetrafluoroethylene microporous support, a polyethylene microporous support, a polypropylene microporous support, a polyaramid microporous support, a polyphenylene sulfide microporous support, a polybenzimidazole microporous support, a polyaryletherketone microporous support, a cellulous microporous support, a polyetherimide microporous support, a polyimide microporous support, a ceramic microporous support, a zeolite microporous support, or a metal microporous support. One particular example of a microporous support that may be used in methods according to the present disclosure is a microporous polysulfone or polyethersulfone support, such as an ultrafiltration membrane.

The first and second precursors may be alternately and repetitively deposited for at least 10 cycles, such as for a total of from 10 to 1000 cycles, preferably from about 20 to about 500 cycles, or about 30 to about 200 cycles, for example for about 50 cycles. As discussed above, it is not necessary for the specific compounds to be the same in each deposition of first membrane precursors, or in each deposition of second membrane precursors. Accordingly, 500 deposition cycles may include 200 depositions using TMC as a first membrane precursor, 200 depositions using TPC as a first membrane precursor, 100 depositions using a mixture of TMC and TPC as a first membrane precursor, and 500 depositions using mPD as a second membrane precursor, where each layer of mPD is deposited on a layer of TMC, TPC, or a mixture of TMC and TPC.

The first and second precursors may be alternately and repetitively deposited for sufficient cycles to form a thin organic film having a thickness of from about 2 nm to about 5,000 nm, such as from about 5 nm to about 500 nm, for example from about 10 nm to about 50 nm. Membranes that are to be used in an RO or NF system may have a thickness from about 10 to about 200 nm, and in particular examples from about 10 nm to about 50 nm. Membranes that are to be used in a gas-separation system may have a thickness from about 100 nm to about 5,000 nm. Increasing the thickness of a membrane reduces the membrane flux but reduces the risk of generating a membrane having a defect.

The first and second precursors are reactive with each other, and each precursor has two or more reactive functional groups for reacting with the other precursor. The reactive functional groups for the first and second membrane precursors may be amines and amine-reactive functional groups, or hydroxyl and hydroxyl-reactive functional groups. For example, the first or second thin film composite membrane precursor may include a polyamine compound； and the second or first, respectively, thin film composite membrane precursor may include a compound having a plurality of independently selected amine-reactive functional groups. Each of the amine functional groups on the polyamine compound may be the same or different from other amine functional groups. Each of the hydroxyl functional groups on a polyol compound may be the same or different from other hydroxyl functional groups.

The average number of reactive functional groups on a precursor may be determined by the weighted average of the number of functional groups in a mixture of precursors. For example, the first precursor may include a 50: 50 mixture of TMC (three amine-reactive functional groups) and TPC (two amine-reactive functional groups) . Such a mixture would have, on average, 2.5 amine-reactive functional groups per precursor. A 75: 25 mixture of a tri-amine and a di-amine would have on average 2.75 functional groups per precursor. Increasing the average number of reactive functional groups per precursor increases the crosslinking in the membrane. Increasing the crosslinking in NF or RO membranes may improve one or more characteristics of the membrane, such as the salt rejection. In examples where an NF or RO membrane is being made, adesirable amount of crosslinking may be achieved when the sum of the average numbers of reactive functional groups for the first and the second precursors is about 5. Less crosslinking is acceptable in gas-separation membranes. In examples where a gas-separation membrane is being made, the sum of the average numbers of reactive functional groups for the first and the second precursors may be from about 2 (non-crosslinked polymer) to 5 (highly crosslinked) .

The polyamine compound may be a diamine compound, such as an aromatic diamine， for example 4， 4’-oxydianiline, or a meta-, or para-phenylene diamine； or an aliphatic diamine, for example ethylene diamine or piperazine. The polyamine compound may be a triamine compound.

Each amine-reactive functional group may be an acyl halide, such as an acyl chloride； asulfonyl halide, such as sulfonyl chloride； or an anhydride. The compound having amine-reactive functional groups may have two independently selected amine-reactive functional groups, such as when the compound is terephthaloyl chloride (TPC) ； isophthaloyl chloride (IPC) ； benzene-1, 3-disulfonyl chloride (BDSC) ； pyromellitic dianhydride (PDMA) ； 1, 5-naphthalenedisulfonyl chloride (NDSC) . The compound may have three or more independently selected functional groups, such as when the compound is: trimesoyl chloride (TMC) ； 1, 3, 6-naphthalenetrisulfonyl chloride (NTSC) ； or 3， 3′， 4， 4′-biphenyltetracarboxylic dianhydride (BPDA) .

When the combination of precursors includes amines and acid chlorides, the resulting polymers include polyamide bonds. When the combination of precursors includes amines and sulfonyl chlorides, the resulting polymer includes polysulfonamide bonds. When the combination of precursors includes amines and anhydrides, the resulting polymer includes polyimide bonds. When the combination of precursors includes hydroxyls and anhydrides, the resulting polymer includes polyesters bonds. When the combination of precursors includes aromatic hydroxyls and aromatic anhydrides, the resulting polymer includes polyarylates.

When the gas-phase deposition requires an elevated temperature to raise the vapor pressure of the precursors, the temperature of the gas-deposition system may be managed to reduce or avoid low temperature cold spots that would allow for precursor condensation or reduced chemical reaction. The temperature may be managed by performing the gas-deposition in an isothermal oven.

Methods according to the present disclosure may perform the gas-deposition using a rotating drum in a spatial molecular layer-by-layer deposition (sMLD) reactor. An exemplary spatial molecular layer-by-layer deposition reactor is discussed in “Low temperature and roll-to-roll spatial atomic layer deposition for flexible electronics “ (P. Poodt, et. al. J. Vac. Sci. Technol. A, Vol. 30, No. 1, 01A142 (2012) ) . This paper generally discusses methods of performing spatial molecular gas-phase deposition on solid, non-porous substrates. Another exemplary spatial molecular layer-by-layer deposition reactor isdiscussed in “Spatial atomic layer deposition on flexible substrates using a modular rotating cylinder reactor” (K. Sharma, et. al. J. Vac. Sci. Technol. A, Vol. 33, No. 1, 01A132-1 (2015) ) . As discussed in the abstract, this paper generally presents results for spatial ALD on flexible substrates using a modular rotating cylinder reactor. The design for this reactor is based on two concentric cylinders. The outer cylinder remains fixed and contains a series of slits. These slits can accept a wide range of modules that attach from the outside. The modules can easily move between the various slit positions and perform precursor dosing, purging, or pumping. The inner cylinder rotates with the flexible substrate and passes underneath the various spatially separated slits in the outer cylinder. These papers are incorporated herein by reference in their entirety.

In another aspect, the present disclosure provides a thin film composite membrane. The membrane may be formed using the method discussed above. The membrane includes a microporous support having thereon molecular layers of (i) one or more first organic compound residues, and (ii) one or more second organic compound residues. Substantially all of the organic compound residues in each molecular layer are covalently bonded to one or more organic compound residues in one or more adjacent molecular layers.

For RO membranes made of mPD/TMC precursors, the membrane may have a crosslink density of at least 75％, such as from about 80％to about 100％, for example about 90％. Crosslink density may be determined as discussed in “Thin film composite polyamide membranes: parametric study on the influence of synthesis conditions” ， RSC Adv. (2015) 5, 54985-97.

The compound residues in any given layer may be a mixture of different residues, and the compound residues in any two layers may be the same or different. For example: the first organic compound residues in a molecular layer may include a plurality of different first organic compound residues； the second organic compound residues in a molecular layer may include a plurality of different second organic compound residues； the membrane may include layers of different first organic compound residues； the membrane may include layers of different second organic compound residues； or any combination thereof.

A compound residue may be bonded to a residue in an adjacent layer via an amide bond, or a sulfonamide bond. For example, abond formed from the reaction between (i) a primary or secondary amine, and (ii) an acid halide such as an acid chloride, a sulfonyl halide such as a sulfonyl chloride, or an anhydride.

In a specific example, the membrane includes: a plurality of molecular layers having compound residues of Formula (I) , and a plurality of molecular layers having compound residues of Formula (II) :

The layers of the different residues alternate so that amide bonds are present between the residues of Formula (I) and the residues of Formula (II) . In this manner, the compound residues in each molecular layer are covalently bonded to the residues in one or more adjacent layers.

In another specific example, the membrane includes a plurality of molecular layers having a mixture of compound residues of Formula (I) and Formula (III) , and a plurality of molecular layers having compound residues of Formula (II) :

In another specific example, the membrane includes a plurality of molecular layers having compound residues of Formula (I) , a plurality of molecular layers having compound residues of Formula (III) , and a plurality of molecular layers having compound residues of Formula (II) . In such an exemplary membrane, it would be understood that each layer of compound residues of Formula (II) is sandwiched between two layers of residues of Formula (I) , sandwiched between two layers of residues of Formula (III) , or sandwiched between a layer of residues of Formula (I) and a layer of residues of Formula (III) . In this manner, the compound residues in each molecular layer are covalently bonded to the residues in one or more adjacent layers.

Other examples of compound residues include:

The microporous support may be a polysulfone microporous support, a polyethersulfone microporous support, a polyvinylidene fluoride microporous support, a nylon microporous support, a polycarbonate microporous support, a polyacrylonitrile microporous support, a polytetrafluoroethylene microporous support, a polyethylene microporous support, a polypropylene microporous support, a polyaramid microporous support, a polyphenylene sulfide microporous support, a polybenzimidazole microporous support, a polyaryletherketone microporous support, a cellulous microporous support, a polyetherimide microporous support, a polyimide microporous support, a ceramic microporous support, a zeolite microporous support, or a metal microporous support.

Examples

Thin film growth on solid substrates. A non-porous metalized polyethylene terephthalate (PET) film was used as a flexible substrate for testing gas-phase molecular layer-by-layer deposition. Titanium coating provides a surface suitable for spectroscopic ellipsometry. During the experiments, the mLBL reactor was located inside an isothermal oven set at 115℃, the drum rotation speed was 20 RPM. 500 A/B cycles were deposited with mPD/TMC precursors. The circumference of the inner cylinder of the rotating drum reactor was 1 meter. The gap between the inner and outer cylinders was 750μm. The operating temperature was 115℃, and the operating pressure was about 1 Torr. Nitrogen was used to provide a gas barrier and to purge excess precursors.

Figure 1 shows the mLBL polyamide film thickness as a function of (Fig. 1a) mPD and (Fig. 1b) TMC flow coefficient (C_v) for mLBL film grown at 115℃ with 500 A/B cycles. This data set shows great self-limiting behavior with respect to the mPD and TMC dosing. Saturation occurs fast and growth does not increase with increased TMC dosing. These data show that the mPD/TMC chemistry has great self-limiting behavior and is truly an mLBL process, as opposed to a chemical vapor deposition (CVD) process.

Figure 2 shows the mLBL polyamide film thickness as a function of drum speed for mLBL film grown at 115℃ with 500 A/B cycles. This data set shows that faster drum rotation speed leads to shorter precursor exposure time, resulting in less deposition per cycle.

Figure 3 shows the polyamide film thickness as a function of the numbers of A/B reaction cycles at 115℃ and 75 rpm drum rotation speed. The data demonstrates the linear growth rate, as expected for the uniform mLbL thin film growth.

Figure 4 shows the IR spectrum of the mLBL thin film made with mPD/TMC precursor at 115℃ with 222 A/B cycles. In the FTIR spectra, three peaks associated with an amide bond appear in the mLbL thin film are at the same wavelengths of the interfacially polymerized polyamide. The remaining peaks associated with the other bonds in the network of mLBL thin film are also similar to those of the spectra of RO thin film made by interfacial polymerization of MPD/TMC. One noticeable difference is the lack of free carboxylic acid peak, indicating that the sMLD thin film are more cross-linked than interfacially polymerized RO thin film, and thus has fewer residual free carboxylic groups.

mLBL films with varying thickness, on both bare glass coverslips and gold-coated glass coverslips, were prepared for surface roughness measurement. For the samples on bare glass coverslips, measurements of film thickness were determined via X-ray reflectivity (XR) and AFM for validation. Figures 5 and 6 show the results from those measurements. Figure 5 shows that the root-mean square (RMS) roughness increases with increasing cycle number. A plot of the ratio of RMS roughness to film thickness illustrates the relative roughness of the film. As shown in Figure 6, the ratio of roughness/thickness actually goes down as the number of mLBL cycles is increased.

Fabrication of thin film composite membranes. Experiments were carried out to fabricate thin film composite membranes by using gas phase molecular layer-by-layer deposition process with either mPD or piperazine as one membrane precursor, and TMC as the other membrane precursor. During the membrane fabrication process, amicroporous polysulfone support was attached to the inner drum of a spatial mLBL reactor, with the non-woven layer facing the inner drum and the microporous layer facing the outer drum so the mLBL thin film grew to form a thin film composite membrane. Unless otherwise specified, all mLBL experiments were carried out at 115℃ and 1 Torr, using a rotating drum reactor having an inner cylinder with a circumference of 1 meter, and a gap between the inner and outer cylinders of 750μm. with the drum rotation speed being in the range of 20-100 RPM.

Tests of the resulting membranes were carried out on composite membranes configured as a flat sheet in a cross-flow test cell apparatus (model CF042, Sterlitech Corp., Kent Washington) with an effective membrane area of 35.68 cm². The test cells were plumbed two in series in each of 6 parallel test lines. Each line of cells was equipped with a valve to turn feed flow on/off and regulate concentrate flow rate, which was set to 1.3 gallon per minute (gpm) in all tests. The test apparatus was equipped with a temperature control system that included a temperature measurement probe, aheat exchanger configured to remove excess heat caused by pumping, and an air-cooled chiller configured to reduce the temperature of the coolant circulated through the heat exchanger.

Composite membranes were first tested with a methyl violet (Sigma-Aldrich) to detect defects. A dye solution comprising 100 ppm methyl violet dye was sprayed on the polyamide surface of the composite membrane and allowed to stand for 1 minute, after which time the dye was rinsed off. Since methyl violet does not stain polyamide, but stains polysulfone strongly, adefect-free membrane should show no dye stain after thorough rinse. On the other hand, dye stain patterns (e.g. blue spots or other irregular dye staining patterns) indicate defects in the composite membranes.

The membranes were cut into 2.55 inch x 6 inch rectangular coupons, and loaded into cross flow test cells. Three coupons (3 replicates) from each type of membranes were tested under the same conditions and the results obtained were averaged to obtain mean performance values and standard deviations. The membrane coupons were first cleaned by circulating water across the membrane in the test cells for 30 minutes to remove any residual chemicals and dyes. Then, synthetic brackish water containing either sodium chloride (NaCl) or sodium sulfate, or magnesium sulfate with conductivity～1000 uS was circulated across membrane at 115 psi and 25℃. After one hour of operation, permeate samples were collected for specified minutes and analyzed.

Solution conductivities and temperatures were measured with a CON 11 conductivity meter (Oakton Instruments) . Conductivities were compensated to measurement at 25℃. The pH was measured with a Russell RL060P portable pH meter (Thermo Electron Corp) . Permeate was collected in a graduated cylinder. The permeate was weighed on a Navigator balance and time intervals were recorded with a Fisher Scientific stopwatch. Permeability, or "A-value" , ofeach membrane was determined at standard temperatures (77 F or 25℃. ) . Permeability is defined as the rate offlowthrough the membrane per unit area per unit pressure. A-values were calculated from permeate weight, collection time, membrane area, and transmembrane pressure. A-values reported herein have units of 10^-5 cm³/s-cm²-atm. Salt concentrations determined from the conductivities of permeate and feed solutions were used to calculate salt rejection values. Conductivities of the permeate and feed solutions were measured, and salt concentrations calculated from the conductivity values, to yield salt rejection values.

Example 1: mPD/TMC mLBL thin film directly deposited on top of a polysulfone microporous support. Apolysulfone (PSU) UF support (MWCO～200K, provided by GE Water) was placed in the rotating drum reactor. The reactorwas equilibrated from 2 hours to overnight before the fabrication of mLBL thin film with mPD/TMC precursors at 115℃, 20 RPM. At the same time a piece of metallized PET film was also coated together with PSU sample. The coating thickness on the metallized PET film was measured by ellipsometry. In this example, the thickness of mLBL PAthin film on the metallized PET was about 100 nm. The performance of the composite membrane sample was tested in～500 ppm NaCl solution in DI water using cross-flow cells underthe conditions as described above. The A-value of this sample was 0.05 and the NaCl rejection was 75％.

Example 2: mPD/TMC mLBL thin film directly deposited on top of a polyether sulfone microporous support. Apolyethersulfone (PES) UF support (MWCO～5K) (purchased from Sterlitech) was placed in the rotating drum reactor. The reactorwas equilibrated from 1.5 hours before the fabrication of mLBL thin film with piperazine/TMC precursors. At the same time a piece of metallized PET film was also coated togetherwith PSU sample. The coating thickness on the metallized PET film was measured by ellipsometry. In this example, the thickness of mLBL coating on the metallized PET film was about 43 nm. The flux and salt rejection was tested on the cross-flow cells using both NaCl and Na2SO4 aqueous solution with conductivity～1000 uS using the testing condition as described earlier. No flux was measured in 40 min.

Example 3: mPD/TMC mLBL thin film directly deposited on top of a cross-linked polyvinyl alcohol treated polysulfone microporous support. A polysulfone UF support (MWCO～200K) was first treated with a water soluble polymer poly (vinyl alcohol) (PVA) to protect the micro-pores of the support at high temperature mLBL conditions, and to prevent mLBL polyamide from growing inside the pores and plugging the pores. PVA (Mw ～146-186,000, 99％mol％hydrolysis, Aldrich) was dissolved in deionized (DI) water to make a 0.075 wt％solution. The PVA solution was then mixed with an organic titanate crosslinker in 1/1 weight ratio.

PVA treatment was conducted using a handframe apparatus consisting of a matched pair of frames in which the porous base membrane could be fixed and subsequently treated with a PVA solution. The following procedure was used. The porous base membrane was first soaked in deionized water for at least 30 minutes. The wet porous base membrane was fixed between two rectangle metal frames (8x11 inch) . The PVA solution poured onto the microporous polysulfone UF support which was enclosed with the metal handframe. After a period of 1 minute, the PVA solution was removed from the surface of the porous base membrane by tilting the assembly comprising the frame and the treated porous base membrane until only isolated drops of the aqueous solution could be observed on the surface of the treated porous base membrane. The treated surface was further treated by exposure to a gentle stream of air to remove isolated drops of the PVA solution. The treated assembly was then placed in a drying oven and maintained at a temperature of 70℃ for a period of about 4 minutes after which the composite membrane was left at room temperature for overnight before testing.

Several samples were made use this protocol. The average A-value of several pieces of the PVA treated PSU was about 50. Several other pieces of cross-linked PVA coated PSU were placed inside the mLBL reactor. The reactor was equilibrated from 2 hours before the fabrication of mLBL thin film with MPD/TMC precursors at 115℃ on top of the PVA treated microporous polysulfone. At the same time a piece of metallized PET film was also coated together with PSU sample. The coating thickness on the metallized PET film was measured by ellipsometry. In this example, the thickness of mLBL coating on the metallized PET was about 100 nm. The performance of the sample was tested in～500 ppm NaCl solution in DI water using cross-flow cells under the conditions described in the previous section. The measured A-value of this sample was less than 0.08 and the NaCl rejection was 88％.

Example 4: mPD/TMC mLBL thin film directly deposited on top of an aluminum oxide treated polysulfone microporous support. Two polysulfone UF (MWCO ～200K) samples were first coated with 10 and 33 nm of Al₂O₃, respectively, using an ALD reactor at a reaction temperature of 50℃, then mLBL was conducted on the Al₂O₃coated PSU supports using MPD/TMC chemistry. The mLBL thickness of both samples was 80 nm.

The samples were flushed with 0.01 M NaOH aqueous solution and then rinsed with DI water thoroughly. The average A-value of these samples was about 3 and the NaCl rejection was 0％. The test was done in a dead-end test cell (HP4750X Stirred Cell, Sterilitech) at 115 psi pressure.

Example 5: mPD/TMC mLBL thin film directly deposited on top of a PEI/TMC treated polysulfone microporous support. A polysulfone UF support (MWCO～200K) was first treated with 1 layer of polyelectrolytes, cross-linked polyethylene imide (PEI) using the following procedure. First, the PEI (Mw～750 K, Aldrich) was dissolved in DI water to make 0.075 wt％solution and TMC was dissolved in Isopar-G to make a 1 wt％solution. Next, 100ml PEI solution was poured onto PSU UF support using a handframe to cover the whole surface for 1 minute, the solution was drained before the support was rinsed with deionized water 2 times. Next, the PSU UF support was rinsed with Isopar-G before applying TMC solution followed by 2 more rounds of Isopar-G rinsing. Finally, the PEI/TMC treated polysulfone UF support was dried in an oven at 70℃ for 4 min and then left at RT for overnight drying. The average A-value of several pieces of the cross-linked PEI/TMC treated PSU was about 74. mLBL was conducted for the PEI/TMC treated polysulfone UF support at 115℃ using MPD/TMC chemistry. The mLBL coating thickness was about 100 nm. The membrane performance tests were performed with～500 ppm NaCl solution in DI water. The measured A-value was about 0.1 with the measured NaCl rejection was 40％.

Additionally, 1 A/B layer of PEI/PAA (poly (acrylic acid) ) was coated on top of the microporous polysulfone support (MWCO＝200,000) . The PEI/PAA coated PSU were found to have an A-value<4, and thus, no mLBL experiments were conducted with the PEI/PAA treated PSU UF support.

Example 6: PIP/TMC mLBL thin film directly deposited on top of an ionic liquid treated polysulfone microporous support. 1-Butyl-1-methylpyrrolidinium bis (trifluoromethanesulfonyl) imide (PYR14-TFSI, CAS Number：223437-11-4 Product Number： B2851) was purchased from TCI America. The following ionic liquid treatment procedure was conducted: 1) lay flat a PSU UF support onto a piece of clean PET sheet with tapes； 2) rinse PSU UF support with isopropanol for 1 min until it was thoroughly wetted； 3) air dry PSU UF support for 5-10 minutes, then place the sample on an automatic thin film applicator； 4) place a doctor blade (with a 2 mil gap) on top of PSU, and then apply about 9 ml ionic liquid onto the area enclosed by using the square-shape doctor blade (4x4 inch) , allow the liquid to totally wet the enclosed area of PSU UF support； 5) push the doctor blade to treat the whole piece of PSU FU support (～4x12” ) at a 2.5 mm/s speed； 6) leave the ionic liquid treated sample in a beaker and let the sample stand up for overnight to allow the excess ionic liquid to drain to the bottom of the sample. After overnight, no excess solution was seen on the surface but the membrane remains wet. The bottom of the sample with visible liquid layer was discarded. The rest of the sample was shipped from Niskayuna, New York to Boulder, Colorado for mLBL coating.

The PYR14-TFSI treated polysulfone UF support membranes were placed in a rotating drum reactor. The reactor was equilibrated from 2 hours to overnight before the fabrication of mLBL thin film with PIP/TMC precursors at 115℃, 20 RPM. At the same time a piece of metallized PET film was also coated together with PSU sample. The coating thickness on the metallized PET film was measured by ellipsometry. In this example, the thickness of mLBL coating was about 22 nm. The samples were soaked in isopropanol for 5 minutes to remove ionic liquid, and rinsed for 2 times, then rinsed with DI water for 3 times to remove isopropanol before performance tests. The membrane performance tests were carried out using a 500 ppm NaCl and a 500 ppm MgSO₄ aqueous solution in DI water using the cross-flow test procedure described previously. The A-value was about 10 and the salt rejections were about 92％and 63％for MgSO₄ and NaCl, respectively.

Example 7: PIP/TMC mLBL thin film directly deposited on top of an ionic liquid treated polysulfone microporous support. Another ionic liquid filled sample was prepared use the same procedure as described in Example 6 except 1) after casting, the sample was left standing up in a beaker for 30 min before shipping from Niskayuna, New York to Boulder, Colorado for mLBL coating； 2) the mLBL coating thickness was about 35 nm. The samples were soaked in isopropanol for 5 minutes to remove ionic liquid, and rinsed for 2 times, then rinsed with DI water for 3 times to remove isopropanol before performance tests. The membrane performance tests were also carried using the same protocol as described in Example 6. The measured A-value was about 10, and the measured salt rejections were 89％and 66％for MgSO₄ and NaCl, respectively.

Examples 1-5 may be viewed as comparative examples with respect to Examples 6 and 7, showing the benefits of using an Ionic Liquid to at least partially fill at least some of the pores of the microporous support under certain process situations (such as when a PSU microporous support is being used at an operating temperature of 115℃ and an operating pressure of 1 Torr.

In the preceding description, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the examples. However, it will be apparent to one skilled in the art that these specific details are not required. Accordingly, what has been described is merely illustrative of the application of the described examples and numerous modifications and variations are possible in light of the above teachings.

Since the above description provides examples, it will be appreciated that modifications and variations can be effected to the particular examples by those of skill in the art. Accordingly, the scope of the claims should not be limited by the particular examples set forth herein, but should be construed in a manner consistent with the specification as a whole.

Claims

A method comprising：

repetitively depositing on to a microporous support， via a gas phase： (i) one or more first thin film composite membrane precursors， and (ii) one or more second thin film composite membrane precursors，

wherein the one or more first membrane precursors are reactive with the one or more second membrane precursors，

wherein the gas-phase deposition is performed isothermally at a temperature that is at or below the lowest thermal decomposition temperature of the first and second membrane precursors， and

wherein the first and second membrane precursors have vapor pressures 0.1 Torr or higher at the temperature and pressure of the gas-phase deposition.
The method according to claim 1， wherein the one or more first thin film composite membrane precursors is a mixture of a plurality of precursors.
The method according to claim 1， wherein the one or more first thin film composite membrane precursors is a plurality of precursors， and the method comprises separately depositing at least one of the plurality of first thin film composite membrane precursors.
The method according to any one of claims 1-3， wherein the one or more second thin film composite membrane precursors is a mixture of a plurality of precursors.
The method according to any one of claims 1-3， wherein the one or more second thin film composite membrane precursors is a plurality of precursors， and the method comprises separately depositing at least one of the plurality of second thin film composite membrane precursors.
The method according to any one of claims 1-5， wherein the gas-phase deposition is performed at a pressure of at least 0.01 Torr， such as from about 0.01 Torr to about 760 Torr， for example from about 0.05 Torr to about 100 Torr， or from about 0.1 to about 10 Torr.
The method according to any one of claims 1-6， wherein the gas-phase deposition is performed at a temperature from about 30 ℃ to about 250 ℃， such as from about 60 ℃ to about 200 ℃， or from about 80 ℃ to about 150 ℃.
The method according to any one of claims 1-7， wherein at least 50％of the pores of the microporous support are at least partially full of a liquid that has， at the gas-phase deposition temperature， a vapor pressure at least two orders of magnitude lower than the gas-phase deposition pressure.
The method according to claim 8， wherein at least 80％， and preferably about 100％， of the pores of the microporous are at least partially full， and preferably substantially completely full， of the liquid.
The method according to claim 8 or 9， wherein the liquid is an ionic liquid that lacks functional groups reactive with the membrane precursors， for example 1-Butyl-1-methylpyrrolidinium bis (trifluoromethanesulfonyl) imide.
The method according to any one of claims 1-10， wherein the one or more first and one or more second precursors are alternately and repetitively deposited for a total of at least 10 cycles， such as for a total of from 10 to 1000 cycles， preferably from about 20 to about 500 cycles， or about 30 to about 200 cycles， for example for about 50 cycles.
The method according to any one of claims 1-10， wherein the first and second precursors are alternately and repetitively deposited to form a thin organic film having a thickness of from about 2 nm to about 5， 000 nm， such as from about 5 nm to about 500 nm， for example from about 10 nm to about 50 nm.
The method according to any one of claims 1-12， wherein the microporous support is a glassy microporous polymeric support having a Tg at least 20℃ higher， such as at least 60℃ or 80℃ higher， than the gas-phase deposition temperature； or is a semi-crystalline microporous polymeric support having a Tm at least 20℃ higher， such as at least 60℃ or 80℃ higher， than the gas-phase deposition temperature.
The method according to any one of claims 1-13， wherein the microporous support is a polysulfone microporous support， a polyethersulfone microporous support， a polyvinylidene fluoride microporous support， a nylon microporous support， a polycarbonate microporous support， a polyacrylonitrile microporous support， a polytetrafluoroethylene microporous support， a polyethylene microporous support， a polypropylene microporous support， a polyaramid microporous support， a polyphenylene sulfide microporous support， a polybenzimidazole microporous support， a polyaryletherketone microporous support， a cellulous microporous support， a polyetherimide microporous support， a polyimide microporous support， a ceramic microporous support， a zeolite microporous support， or a metal microporous support.
The method according to any one of claims 1-14， wherein：

the one or more first thin film composite membrane precursors include a polyamine compound， and

the one or more second thin film composite membrane precursors include a compound having a plurality of independently selected amine-reactive functional groups.
The method according to claim 15， wherein the polyamine compound is a diamine compound or a triamine compound.
The method according to claim 16， wherein the diamine compound is： an aromatic diamine， such as 4， 4’ -oxydianiline， or a meta-， or para-phenylene diamine； or an aliphatic diamine， such as ethylene diamine or piperazine.
The method according to any one of claims 15-17， wherein each amine-reactive functional group is an acyl halide， such as an acyl chloride； a sulfonyl halide， such as sulfonyl chloride； or an anhydride.
The method according to any one of claims 15-18， wherein the compound has two independently selected functional groups， such as when the compound is terephthaloyl chloride (TPC) ； isophthaloyl chloride (IPC) ； benzene-1， 3-disulfonyl chloride (BDSC) ； pyromellitic dianhydride (PDMA) ； 1， 5-naphthalenedisulfonyl chloride (NDSC) .
The method according to any one of claims 15-18， wherein the compound has three or more independently selected functional groups， such as when the compound is： trimesoyl chloride (TMC) ； 1， 3， 6-naphthalenetrisulfonyl chloride (NTSC) ； or 3， 3′， 4， 4′-biphenyltetracarboxylic dianhydride (BPDA) .
The method according to any one of claims 1-20， wherein the gas-deposition is performed in an isothermal oven.
The method according to any one of claims 1-21， wherein the gas-deposition is performed on a rotating drum in a spatial molecular layer-by-layer deposition reactor.
A thin film composite membrane comprising：

a microporous support having thereon molecular layers of (i) one or more first organic compound residues， and (ii) one or more second organic compound residues， wherein substantially all of the organic compound residues in each molecular layer are covalently bonded to one or more organic compound residues in one or more adjacent molecular layers.
The thin film composite membrane according to claim 23， wherein the molecular layers of the first organic compound residues comprise a plurality of different first organic compound residues.
The thin film composite membrane according to claim 23， wherein the membrane comprises layers of different first organic compound residues.
The thin film composite membrane according to any one of claims 23-25， wherein the molecular layers of the second organic compound residues comprise a plurality of different second organic compound residues.
The thin film composite membrane according to any one of claims 23-25， wherein the membrane comprises layers of different second organic compound residues.
The thin film composite membrane according to any one of claims 23-27， wherein the microporous support is a polysulfone microporous support， a polyethersulfone microporous support， a polyvinylidene fluoride microporous support， a nylon microporous support， a polycarbonate microporous support， a polyacrylonitrile microporous support， a polytetrafluoroethylene microporous support， a polyethylene microporous support， a polypropylene microporous support， a polyaramid microporous support， a polyphenylene sulfide microporous support， a polybenzimidazole microporous support， a polyaryletherketone microporous support， a cellulous microporous support， a polyetherimide microporous support， a polyimide microporous support， a ceramic microporous support， a zeolite microporous support， or a metal microporous support.
The thin film composite membrane according to claim 23， wherein the first and second organic compound residues include the reaction products between m-phenylene diamine and trimesoyl chloride， and the membrane has a crosslinking density of at least 75％.