Environmental Remediation with Functional Aerogels and Xerogels

Abstract

Several different types of aerogels and xerogels are demonstrated as effective sorbents for the capture and/or immobilization of radionuclides and other contaminants in gaseous form [e.g., Hg_(g), I_2(g), Xe, Kr] as well as ionic form (e.g., Cd²⁺, Ce⁴⁺, Cs⁺, Cu²⁺, Fe²⁺, Hg²⁺, I⁻, IO₃ ⁻, Kr, Pb²⁺, Rb⁺, Sr²⁺, ⁹⁹Tc⁷⁺, U⁶⁺, Zn²⁺). These sorbents have unique properties, which include high specific surface areas, high pore volumes, a range of pore sizes, and functionalities that provide methods for binding radionuclides and other contaminants, generally through physisorption, chemisorption, or a combination thereof. This combination of properties and functionalities makes these types of materials ideal for use as sorbents for capturing radionuclides. The primary base materials that will be discussed in this paper include Ag⁰‐functionalized silica aerogels, Ag⁺‐impregnated aluminosilicate aerogels, Ag⁰‐functionalized aluminosilicate aerogels, metal‐impregnated (non‐Ag) aluminosilicate aerogels and xerogels, sulfide‐based aerogels, and carbon‐based aerogel composites. For the capture of I_2(g), the materials reported herein show some of the highest iodine loadings ever reported for inorganic sorbents. For the capture of ionic species, these materials also show promise as next‐generation materials for active radionuclide remediation. This progress report describes materials fabrication, general properties, and environmental remediation applications.

Functionalized aerogels and xerogels are demonstrated for use in the removal of contaminants from aqueous and gaseous streams, including species such as halides, noble gases, actinides, lanthanides, alkalis, and alkaline earths that could pose environmental and health concerns if not captured and contained properly. This paper provides a progress report of the current state of the art in this area.

1. Introduction

Porous sorbents including aerogels, xerogels, metal–organic frameworks (MOFs), and zeolites have received a lot of attention over the past several decades for environmental remediation applications. These sorbents have a variety of uses including the purification of gaseous and aqueous streams through chemisorption, physisorption, ion exchange, molecular sieving, and combinations thereof. All these materials have high specific surface areas (SSA), which are measured in terms of surface area per mass of material (i.e., m² g⁻¹), with values ranging from a few hundred m² g⁻¹ (e.g., zeolites) to several thousand m² g⁻¹ (e.g., MOFs). Table 1 presents examples of some SSA values and pore volumes (V _p) for different porous materials in these families.^{[ 1} , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ^{10 ]} To put this into perspective, some MOFs have more surface area of an American football field (5350 m²) per gram of material, e.g., DUT‐32 with 6411 m² g⁻¹.^{[ 11 ]}

Table 1.

Summary of specific surface areas (SSA) and pore volumes (V _p) from the literature for a variety of aerogels, xerogels, chalcogels, metal–organic frameworks (MOFs), and zeolites. Materials are listed in descending order by material class in terms of SSA values

Material class	Material	SSA [m2 g−1]	V p [cm3 g−1]	Ref.
Aerogel	Carbon aerogel	1040–1920	0.7–1.9	[ 3 ]
	Silica aerogel	820–1110	2.6–4.7	[ 1 ]
	Al‐Si‐O	940	2.4	[ 2 ]
	Na‐Al‐Si‐O	240–880	1.3–3.3	[ 2 ]
	Resorcinol‐formaldehyde	360–830	0.4–3.1	[ 3 ]
Xerogel	Carbon xerogel	600–850	1.4–1.6	[ 5 ]
	Silica xerogel	420–970	0.2–1.2	[ 4 ]
	Sn2S3	0.14	–	[ 6 ]
Chalcogel	Sn2S3 chalcogel	460	3.9	[ 6 ]
	PtGe2S5 chalcogel	420	3.1	[ 6 ]
MOF	DUT‐32	6411	2.91	[ 11 ]
	MOF‐5	2890	1.2	[ 7 ]
	ZIF‐8	1080	0.3	[ 8 ]
Zeolite	Mordenite	570	0.3	[ 10 ]
	ZSM‐5	410	0.1	[ 10 ]
	Zeolite A	120–240	0.2–0.5	[ 9 ]

The SSA parameter may be used to compare porous sorbents by normalizing the values on a mass basis, but it does not account for porous materials that are assembled using high‐density moieties containing heavy metals such as those used in fabricating chalcogen‐based aerogels (e.g., Ge₄S₁₀ ⁴⁻, SnS₄ ⁴⁻, SnSe₄ ⁴⁻), called chalcogels.^{[ 12 ]} Thus, in some cases, a correction factor can be used to renormalize the SSA values as a silica equivalent (SSA _eq) value using Equation (1) where SSA _m is the measured specific surface area for the material of interest, $m{w}_{{\text{SiO}}_2}$ is the molecular weight of SiO₂, and mw _x is the molecular weight of the material of interest normalized to two anions.^{[ 13} , ^{14 ]} As an example, PtGe₂S₅ (SSA _m = 491 m² g⁻¹)^{[ 15 ]} would be normalized to Pt_0.4Ge_0.8S₂ where mw _x = 200.3 g mol⁻¹ and the corresponding SSA _eq is 1636 m² g⁻¹

$$SS{A}_{\text{eq}}=SS{A}_{\text{m}}\text{*}\left(m{w}_{\text{x}}\text{/}m{w}_{{\text{SiO}}_2}\right)$$ (1)

While previous reviews^{[ 16} , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ^{21 ]} have shown the applications where different porous materials can be used to capture and/or immobilize various contaminants, this paper focuses solely on functionalized aerogels and xerogels. Materials discussed herein include oxide‐based and sulfide‐based aerogels and xerogels as well as carbon‐based aerogel composites that have been demonstrated to capture a range of environmental contaminants including those in gaseous form [e.g., Hg_(g), I_2(g), Xe, Kr] as well as some in ionic form (e.g., Cd²⁺, Ce⁴⁺, Cs⁺, Cu²⁺, Fe²⁺, Hg²⁺, I⁻, IO₃ ⁻, Kr, Pb²⁺, Rb⁺, Sr²⁺, ⁹⁹Tc⁷⁺, U⁶⁺, Zn²⁺) (see Figure 1 ). Descriptions of common synthesis methods utilized in the literature as well as some precursor selection guidelines will be provided for each class of sorbent.

Figure 1.

Overview of the material classes and some target contaminants discussed within the context of this progress report.

2. Synthesis Methods

Methods for producing aerogels and xerogels vary widely based on the target chemistry, but the processes involved are similar—some of these methods are briefly described in the process flow diagram shown in Figure 2a–g. To start this process, the target chemistry must be selected. If the goal is to reach an oxide‐based chemistry, alkoxides are a typical selection of precursor, e.g., tetraethyl orthosilicate (TEOS) for building an Si—O network, described in Section 2.1. For building a chalcogen‐based gel network, an entirely different set of precursor chemicals must be used, described in Section 2.2. In both systems, a compatible solvent is used as a medium in which the reactions take place; in some cases, it is also used to dissolve precursor solids. However, for both oxide‐based and chalcogen‐based sol‐gel systems, precursors undergo a similar series of reactions (Figure 2h–l)^{[ 22} , ^{23 ]} that result in interlinking chemical bonds between the precursors to create a sol, which is a series of colloidal particles in solution (Figure 2c). This mixture can be placed inside a container that will ultimately become the final shape of the gel in subsequent phases of the process. Once the sol is generated, a gel can form, but the gelation time can vary from minutes to months, depending on the chemistry of the precursors, solution pH, temperature, and the presence of a catalyst. The container selection is also important to make sure that the gel can be removed from the container at later stages without damaging the mechanical integrity of the product.

Figure 2.

Process flow diagram for making g) aerogels or f‐2) xerogels starting with a) precursor solutions; additional stages include b) mixing precursors, c) the sol, d) hydrogels, e) alcogels, and f‐1) liquid CO₂ exchange; Adapted with permission.^{[ 2 ]} Copyright 2017, American Chemical Society. h) Hydrolysis/esterification, i) alcohol condensation/alcoholysis, and j) water condensation/hydrolysis processes where “R” denotes a charged organic group, e.g., CH₃ ⁺, C₂H₅ ⁺ (after Brinker^{[ 22 ]}). Adapted with permission.^{[ 2 ]} Copyright 2017, American Chemical Society. k) Thiolysis and l) condensation of Ge(OR)₄ metal alkoxide using H₂S where “R” denotes a charged organic group, e.g., CH₃ ⁺, C₂H₅ ⁺ (after Brock and Yu^{[ 23 ]}).

Once gelation occurs, if water is the medium, the product is called a hydrogel (Figure 2d). At this stage, it is common to allow the gel to remain in the medium for an extended time, called aging. Aging is a necessary step to allow for completion of hydrolysis and polycondensation reactions and to allow the gel to become more rigid. Once this process is complete, where the time varies for each gel chemistry, the reaction byproducts need to be removed through passive diffusion from the liquid medium within the gel matrix through a series of solvent exchanges. This typically involves water, such as 50/50 ethanol/water for oxide‐based chemistries or 50/50 formamide/water for chalcogel chemistries. The more times this process is conducted, the more effectively byproducts are removed, but additional rinses will increase the volume of waste produced. The larger the gel pieces are during this step, the more difficult this phase becomes.

Following this stage, the water must be removed though a series of exchanges with an organic solvent; typically, this solvent is ethanol. Again, this is accomplished through passive diffusion to dilute the water present within the gel matrix, and mixing is recommended during this step to help agitate the gels. If the solvent is completely exchanged to an alcohol, this product is called an alcogel (Figure 2e). At this stage, the sample can be dried (e.g., in air, under vacuum) to create a xerogel (Figure 2f‐2); this process can be destructive to the gel network and often results in significantly collapsed pore structures and reduced SSA values. Some things can be done to prevent pore structure collapse and that will be discussed in a later section. An alternative technique is to solvent exchange an alcogel to a polar organic solvent, such as hexane, prior to drying to reduce capillary forces.^{[ 24 ]}

If the alcogel is then solvent exchanged with liquid CO₂ in an autoclave to remove the alcohol (Figure 2f) using several rinses through passive diffusion, the CO₂ can be taken to a supercritical state (i.e., ≥7.39 MPa and ≥31.1 °C) and vented slowly as a gas to produce an aerogel (Figure 2g). This process of allowing the CO₂ to enter a supercritical state significantly reduces the surface energy and prevents the gel network from collapsing during this transition, allowing the gel to retain the SSA present within the alcogel phase. Examples of tools to aid in the processes of solvent exchange and critical point drying are shown in Figure S1 (Supporting Information).

2.1. Making Oxide‐Based Aerogels

2.1.1. Precursor Selection and Target Reactions

Table 2 lists a series of different species that can be used to fabricate oxide gels and/or for usage in sol‐gel techniques that sometimes yield non‐gel products.^{[ 2} , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ^{41 ]} Some of these reactants are liquids at room temperature, some are solids, and some are dissolved (or suspended) in a secondary liquid, which is typically an alcohol (e.g., methanol, ethanol, isopropanol). The hydrolysis rates of metal alkoxides vary widely where some compounds are quite stable in air (e.g., TEOS) while others can hydrolyze readily in air [e.g., aluminum tri‐sec‐butoxide or Al(OBu^s)₃, sodium ethoxide or NaOEt] and require storage and handling in an inert atmosphere such as a nitrogen glovebox. Depending on the target chemistry, sometimes several options are available for a given bond addition such as tetraethyl orthosilicate (TEOS) or tetramethyl orthosilicate (TMOS) for adding Si—O moieties to an oxide‐gel network.

Table 2.

Summary of various alkoxide precursor options and chemistries that they provide for the gel network, sorted alphabetically based on network cation charge (M ^{n +}) from high (n = 5) to low (n = 1) oxidation states. Information provided here include the bond addition to the network, the additive to achieve this species, common abbreviations used for these additives in the literature, and references where these materials have been used to fabricate gel materials or as additives in sol‐gel processes

M n + (n = )	Network cation	Bond addition(s)	Additive (formula)	Common (or IUPAC) name	Common abbrev.	Ref.
+5	Nb5+	Nb—O	Nb(OC2H5)5	Niobium(V) ethoxide	NbOEt	[ 25 ]
	Ta5+	Ta—O	Ta(OC2H5)5	Tantalum(V) ethoxide	TaOEt	[ 26 ]
	V5+	O=V—O	VO(OC3H7)3	Vanadyl isopropoxide	VOPr	[ 27 ]
+4	Ge4+	Ge—O	Ge(OC2H5)4	Tetraethyl orthogermanate	GeOEt	[ 28 ]
	Si4+	Si—O	Si(OC2H5)4	Tetraethyl orthosilicate	TEOS	[ 2 ]
			Si(OCH3)4	Tetramethyl orthosilicate	TMOS	[ 29 ]
	Sn4+	Sn—O	Sn(OC3H7)4	Tin(IV) isopropoxide	Sn(IV)OPr	[ 30 ]
	Ti4+	Ti—O	Ti(OC2H5)4	Titanium(IV) ethoxide	TiOEt	[ 31 ]
	Zr4+	Zr—O	Zr(OC3H7)4	Zirconium(IV) propoxide	ZrOPr	[ 32 ]
+3	Al3+	Al—O	Al(OC4H9)3	Aluminum tri‐sec‐butoxide	Al(OBus)3	[ 2 ]
	B3+	B—O	B(OC2H5)3	Triethyl borate	TEOB	[ 33 ]
			B(OCH3)3	Trimethyl borate	TMOB	[ 34 ]
	Sb3+	Sb—O	Sb(OC2H5)3	Antimony(III) ethoxide	SbOEt	[ 35 ]
	Y3+	Y—O	Y(OC3H7)3	Yttrium(III) tris(isopropoxide)	YOPr	[ 36 ]
+2	Ca2+	Ca—O	Ca(C3H7O2)2	Calcium 2‐methoxyethoxide	CaOMeEt	[ 37 ]
	Mg2+	Mg—O	Mg(OCH3)2	Magnesium methoxide	MgOMe	[ 38 ]
	Cu2+	Cu—O	Cu(OCH3)2	Copper(II) methoxide	CuOMe	[ 39 ]
+1	K+	K—O	KOC2H5	Potassium ethoxide	KOEt	[ 40 ]
			KOCH3	Potassium methoxide	KOMe	[ 41 ]
	Na+	Na—O	NaOC2H5	Sodium ethoxide	NaOEt	[ 2 ]
			NaOCH3	Sodium methoxide	NaOMe	[ 2 ]

The reactions that occur in the addition of water to an alkoxide are summarized in Figure 2h–j for a Si‐based alkoxide based on work by Brinker.^{[ 22 ]} These reactions include hydrolysis, alcohol condensation, water condensation, alcoholysis, and esterification. The differences between these reactions include the bonds that are formed and the byproducts that are released as a result.

Catalyst additions to these processes can be used to speed up the gelation time by accelerating the hydrolysis and polycondensation reactions. An array of catalysts have been reported for oxide gels from acids (e.g., HCl, H₂SO₄, HF, HNO₃, and acetic acid or HOAc) to bases (e.g., KOH, NH₄OH), as well as salts (e.g., KF).^{[ 22} , ^{42 ]} A good example of the differences in gelation times for uncatalyzed reactions versus catalyzed reactions (i.e., 0.05:1 catalyst:TEOS, by mole) was provided by Pope and Mackenzie^{[ 43 ]} for comparing gelation times for TEOS hydrolyzed with 4 moles of water per mole of TEOS. They reported gelation times of 12 h for HF, 72 h for HOAc, 107 h for NH₄OH, or >1000 h if uncatalyzed.^{[ 43 ]} It is important to fully remove the catalyst during solvent exchanging to prevent corrosion of autoclaves and other equipment at subsequent steps in the gel production process.

2.1.2. Loading Getters onto Aerogel Scaffolds

Getters can be added to oxide aerogel scaffolds using a variety of techniques including ion exchange, impregnation, and functionalization. For instance, materials can be soaked in a solution containing a salt with the desired metal cation such as using AgNO₃ for adding Ag⁺,^{[ 44 ]} soaked in a molten salt (e.g., AgNO₃, 50/50 AgNO₃/AgClO₃ by mass),^{[ 45 ]} or by using (3‐mercaptopropyl) trimethoxysilane (3MPTS) to add thiol (—SH) tethers followed by soaking in AgNO₃ (see Figure 3 ).^{[ 46 ]} In the case of thiolation, the Ag links to the —SH tethers that are bound to an organic moiety connected to the silica backbone of the base aerogel material. The thiolation process is conducted by first hydrating the gels in a humid glass jar overnight, adding 3MPTS dropwise to the gels at ≈1.5 mL g⁻¹ of unhydrated sample (the mass before hydration), loading the gels into a high‐pressure autoclave at 150 °C, loading the vessel with liquid CO₂ at 24 MPa, and allowing this to sit for up to 2 d in supercritical CO₂. Following the soak at 150 °C, the CO₂ can be vented as a gas and the gel material removed for the subsequent step of adding the getter (e.g., Ag⁺ with AgNO₃).

Figure 3.

Cartoon showing Ag⁰ moieties attaching to thiol tethers using 3MPTS (see Section 2.1.2). Reproduced with permission.^{[ 16 ]} Copyright 2016, Elsevier.

Ion exchange processes are commonly practiced for replacing cations within zeolite frameworks and this same approach can be applied to aerogel/xerogel chemistries. The selected getter loading process will depend on the gel chemistry, chemical compatibility with the available options, and the getter chemistry of choice. For instance, each getter will have different solubilities in different solvents, where some solvents could destroy (e.g., dissolve) the scaffold network; dissolving the getter additive in the base solvent can drastically change the pH of the solution.^{[ 47 ]} For example, acidic solutions can dissolve silica networks.

2.2. Making Chalcogen‐Based Aerogels

Chalcogenide aerogels, also called chalcogels, can be prepared through different, albeit similar, methods to oxide‐based gels but generally with non‐oxide precursors. These processes include 1) thiolysis of metal alkoxides with H₂S, 2) aggregation of nanoparticles, or 3) metathesis reactions (i.e., chemical linkage of chalcogenide clusters with interlinking metal ions). Several different example chemistries for each synthesis option are provided in Table 3 ,^{[ 12} , ¹³ , ¹⁴ , ¹⁵ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ , ^{62 ]} and additional information can be found in a previous review by Brock and Yu^{[ 23 ]} as well as papers by Krebs^{[ 63 ]} and Kanatzidis et al.^{[ 12} , ⁵⁵ , ⁵⁹ , ^{62 ]} The chemistries available for study are broad with a wide range of applications discussed in later sections. Chalcogel compositions include chemistries of S‐based,^{[ 13 ]} Se‐based,^{[ 64 ]} and Te‐based^{[ 65 ]} networks.

Table 3.

Summary of different types of commonly studied chalcogel chemistries. Note that solvents used are not included herein and the reader is referred to the primary literature for the full synthesis procedures

Method	Chemistry (family)	Primary precursor(s) a)	Interlinking metal(s), M	Ref(s).
Alkoxide thiolation	Ge‐S	Ge(OC2H5)4, H2S	Ge4+	[ 48 ]
	La‐S	La(NSi2C6H18)3, H2S	La3+	[ 49 ]
	Nb‐S	Nb(OC2H5)5, H2S	Nb5+	[ 50 ]
	Ti‐S	Ti(OC3H7)4, H2S	Ti4+	[ 50 ]
	W‐S	W(OC2H5)6, H2S	W6+	[ 51 ]
	Zn‐S	Zn(OC4H9)2, H2S	Zn2+	[ 52 ]
Nanoparticle aggregation	CdS	Cd(NO3)2, Na2S	Cd2+	[ 53 ]
	CdSe	Cd(NO3)2, Na2Se	Cd2+	[ 53 ]
	PbS	Pb(NO3)2, Na2S	Pb2+	[ 53 ]
	ZnS	Zn(OAc)2, Na2S	Zn2+	[ 53 ]
Metathesis reactions	Pt(Ge,Sn)(S,Se)	Ge4(S,Se)10 4−, Sn4Se10 4−, Sn2Se6 4−, Sn(S,Se)4 4−	Pt2+	[ 13 , 14 , 15 ]
	(Mo,W)‐M‐S	(Mo,W)S4 2−	Co2+, Ni2+	[ 54 , 55 ]
	(K,Na)(Sb,As)‐Fe‐(S,Te)	AsS3 3−, SbS3 3−, SbTe3 3−	Fe2+, K+, Na+	[ 56 , 57 ]
	(Sn,Sb)‐M‐(S,Se)	Sn2(S,Se)6 4−, Sn(S,Se)4 4−, SbSe4 3−	Sn2+, Sb3+	[ 12 , 58 ]
	Zn‐Sn‐S	SnS4 4−, Sn2S6 4−, Sn4S10 4−	Zn2+	[ 59 ]
	Fe‐Sn‐S	Fe4S4 m −, Sn2S6 4−	Fe4S4 m −, Sn2S6 4−	[ 60 ]
	Fe‐M‐Sn‐S	Fe4S4 m −, Sn2S6 4−	Zn2+, Sn2+, Ni2+, Co2+	[ 61 ]
	Mo‐Co‐M‐S	MoS4 2−	Co2+, Pb2+, Cd2+, Pd2+, Cr3+, Bi3+	[ 62 ]

^a)OAc denotes acetate (or CH₃COO⁻).

2.2.1. Alkoxide Thiolysis

The thiolysis synthesis route for making chalcogels is similar to that described in Section 2.1 for producing oxide‐based aerogels using metal alkoxides. In this case, the same metal alkoxides can be used but, instead of adding H₂O to hydrolyze the alkoxides and build an oxide‐based network through —O— bridges, H₂S_(g) is bubbled through the solution; the effect of adding H₂S is similar to that of adding H₂O, but adding H₂S results in —S— bridges. This process is shown schematically in Figure 2k,l, which is based off a description provided by Brock and Yu.^{[ 23 ]}

This same process can also be used to create —Se— or —Te— bridges by exposing materials to H₂Se_(g) or H₂Te_(g), respectively, but these gases are extremely toxic. Alternative methods have been employed by adding metal salts and Se⁰ or Te⁰ to a material followed by heating to precipitate crystals with a matrix such as CdSe or CdTe.^{[ 66 ]}

2.2.2. Nanoparticle Aggregation

For nanoparticle aggregation, metal chalcogenides have been shown to interconnect if prepared through room‐temperature reverse‐micellar strategies or high‐temperature arrested‐precipitation. The chemistries demonstrated with this technology include CdS, CdSe, PbS, and ZnS. It was observed in the work by Mohanan and Arachchige^{[ 53 ]} that these materials tended to have short‐range order evidenced by semicrystalline nanoparticles observed with transmission electron microscopy (TEM). The SSA values for gels produced by Mohanan and Arachchige^{[ 53 ]} with this technique ranged from 120–250 m² g⁻¹.

In addition to this work, chalcogels produced through nanoparticle agglomeration can be ion exchanged to produce even new chemistries. In a study by Yao et al.,^{[ 64 ]} CdSe gels were exchanged with Ag⁺ using AgNO₃ dissolved in methanol to yield Ag₂Se aerogels. The starting CdSe gel had an SSA of 133 m² g⁻¹ and, after exchange, the Ag₂Se aerogel had an SSA of 111 m² g⁻¹.

2.2.3. Metathesis Reactions: Chemical Linkage of Clusters and Interlinking Metals

Producing chalcogels using the chalcogenide clusters and interlinking metal ions is a fairly recent discovery^{[ 13} , ^{14 ]} that opens up more options and chemical flexibility than previous methods for making new chalcogel chemistries ranging from simple to complex. The available list of clusters is still growing as new chemistries can be synthesized following existing procedures. Some of the more well‐studied chemistries include Ge‐(S,Se), Sn‐(S,Se), Mo‐S, and W‐S systems. In most cases, the chalcogenide clusters are not commercially available and must be synthesized. For instance, one of the salts used to provide Ge₄S₁₀ ⁴⁻, [(CH₃)₄N]₄Ge₄S₁₀, is made hydrothermally from (CH₃)₄NOH, Ge powder, and S powder where the product solution can be filtered and a solvent added (e.g., acetone) to induce precipitation/crystallization toward the target compound.^{[ 67 ]} Common solvents for dissolving these precursors include water and formamide. Two examples, which are shown in Figure 4 are the synthesis of CoNiMo₂S₄ and Sn₂S₃ chalcogels described elsewhere.^{[ 6} , ⁵⁵ , ⁶⁸ , ^{69 ]} For CoNiMo₂S₄, separate solutions of Ni(NO₃)₂, CoCl₂, and (NH₄)₂MoS₄ are dissolved in formamide, combined and mixed, and then casted into polypropylene vials to undergo gelation. For Sn₂S₃ chalcogel synthesis, Na₄Sn₂S₆ ·14H₂O and Sn(II) acetate are separately dissolved in formamide, combined and mixed, and then cast to undergo gelation.

Figure 4.

Precursor mixing and casting for making a) CoNiMo₂S₈ and b) Sn₂S₃ chalcogels in formamide followed by gel production after critical point drying (preceded by solvent exchanges, not shown). The graphics for (b) were reproduced with permission.^{[ 68 ]} Copyright 2014, American Chemical Society.

3. Iodine Capture with Metal‐Functionalized Aerogels and Xerogels

3.1. Silver‐Functionalized Silica Aerogel

Silver‐functionalized silica aerogels (Ag⁰SA) are a relatively new class of iodine sorbents that are being evaluated by Matyáš et al.^{[ 46} , ^{70 ]} to replace silver mordenite (Ag⁰Z) in nuclear fuel reprocessing off‐gas schemes proposed by the U.S. Department of Energy Office of Nuclear Energy (DOE‐NE); note that DOE‐NE currently considers Ag⁰Z as the baseline technology for I_2(g) capture. The Ag⁰SA sorbent has documented decontamination factors (DFs) upward of 10000 in laboratory tests of dissolver off‐gas streams, where DF is a metric for determining the effectiveness of the sorbent to remove a contaminant of interest from a stream (sometimes containing other species that might compete with the binding sites).^{[ 46} , ^{71 ]} The DF parameter is defined in Equation (2) where CE is the capture efficiency (e.g., if 99.99% is captured, CE = 0.9999 and DF = 10000).^{[ 16 ]} The process for synthesizing these materials consists of several steps: 1) commercially available silica aerogel granules are heat‐treated to remove hydrophobic silyl functional groups from the aerogel surfaces, 2) a thiol (i.e., —SH) layer is added to the gel using the process mentioned above with 3MPTS (thiolation; see Section 2.1.2), 3) silver ions (i.e., Ag⁺) are added to the gel through chemisorption to the sulfur in the thiol groups (i.e., Ag‐impregnation; —S—Ag), and 4) the silver is reduced under a flowing stream of H₂/Ar gas (i.e., Ag‐functionalization; —SH + Ag⁰; see Figure 3. Upon I_2(g) capture, AgI crystals are formed within the composite sorbent.

$$\text{DF}=1\text{/}\left(1-\text{CE}\right)$$ (2)

Some of the benefits of the Ag⁰SA over the Ag⁰Z include a higher capacity for iodine (i.e., ≈480 mg g⁻¹, ≈3–4× greater than Ag⁰Z)^{[ 46} , ⁷⁰ , ^{71 ]} and a higher functionality after aging in NO_x atmosphere and high‐humidity environments (both of which could be observed in a nuclear fuel reprocessing facility). Additionally, Ag⁰SA has thiol groups that help prevent oxidation of Ag⁰, a process that renders the sorbent less effective, and it can be consolidated into a multiphase waste form containing AgI particles encapsulated within a glassy silica matrix.

In a recent paper by Matyáš et al.,^{[ 70 ]} X‐ray photoelectron spectroscopy analysis on base sorbents and sorbents loaded with I_2(g) revealed some key insights into the way that this sorbent works. For instance, when the sorbents were exposed to a 1 vol% NO_(g) atmosphere during an aging experiment, the Ag⁰ nanoparticle sizes increased and free sulfate was observed on the surfaces of the material due to oxidized thiol groups (i.e., S²⁺ → S⁶⁺). These changes to the material prevented full utilization of the Ag⁰ for I_2(g) capture. In some cases, Ag₂SO₄ was observed, providing evidence that the thiol layers help act as a buffer to prevent oxidation of the Ag⁰, because the Ag₂SO₄ crystals tend to surround the Ag⁰ nanocrystals and act as a barrier to further Ag⁰ oxidation. Unfortunately, this barrier layer lowers the Ag⁰ accessibility during iodine capture experiments, resulting in decreased I_2(g) sorption capacity. Studies demonstrated that the sorbent could be regenerated under a reducing gas stream to revert the oxidized Ag⁺ back to Ag⁰ and oxidized SO₄ ²⁻ back to S²⁺. After aging, the Ag⁰SA showed a reduced I_2(g) sorption capacity of 43 relative%, but this was in comparison to a 85 relative% reduction for the Ag⁰Z,^{[ 72} , ^{73 ]} further accentuating the benefits of the Ag⁰SA sorbent.

Since the initial discovery of these materials in 2010, several improvements have been made over the last decade. One of the most important improvements has been the increased mechanical rigidity, or the reduction in friability, of the base sorbent using heat treatments and additives during the sorbent preparation process (e.g., TEOS). It was discovered that heat treating the silica aerogel granules at 700–800 °C yielded granules with high resistance to damage and the granules retained high SSA (≈950 m² g⁻¹). Another important development is the consolidation of the material into a dense waste form. After iodine loading, these materials can be readily consolidated using a variety of techniques with negligible iodine loss including hot uniaxial pressing, spark plasma sintering, and hot isostatic pressing.^{[ 46} , ⁷⁴ , ^{75 ]}

Finally, in a different study,^{[ 76 ]} the Ag⁰SA sorbents were demonstrated to remove iodate (i.e., IO₃ ⁻) and iodide (i.e., I⁻) from aqueous solutions; iodine can be present as these species in radionuclide‐contaminated groundwater as well as in caustic scrubbing solutions from radiological off‐gases. The efficacy of this sorbent to remove I⁻ was demonstrated in a variety of different solutions including some with competing halide ions Br⁻ and Cl⁻ as well as the removal of IO₃ ⁻ after a reduction to I⁻ was observed. In the presence of Cl⁻, Br⁻, and I⁻, the Ag⁰SA showed preferential removal in the order of I⁻ > Br⁻ > Cl⁻. For I⁻ removal, a distribution coefficient (K _d) of 1.46 × 10⁴ in off‐gas simulant was demonstrated as well as a K _d of 1.5 × 10⁴ in pure deionized water (DIW); K _d is defined in Equation (3) where c _i is the initial concentration of species in solution, c _t is the concentration after a time interval, V is the volume of solution (mL), and m is the mass of solid. Also, the kinetics of I⁻ removal were faster than that of IO₃ ⁻ removal; 98% of I⁻ was removed within 1 h of contact in DIW (c _i = 0.044 mmol L⁻¹) whereas 94% of IO₃ ⁻ was removed after contact for ≈12 d in DIW (first‐order sorption rate = 9.35 × 10⁻³ h⁻¹)

$$K_{\text{d}}=\left(\frac{c_{\text{i}}-c_{\text{t}}}{c_{\text{t}}}\right)\left(\frac{V}{m}\right)$$ (3)

3.2. Aluminosilicate Aerogel and Xerogels for Iodine Capture

3.2.1. Silver‐Loaded Aluminosilicates for Iodine Capture

Silver‐loaded aluminosilicate aerogels were developed as follow‐on work to the silver‐functionalized silica aerogels.^{[ 2 ]} For this work, Na‐Al‐Si‐O and (Na‐free) Al‐Si‐O aerogels were fabricated from metal alkoxides including TEOS, Al(OBu^s)₃, NaOMe, and NaOEt (see Table 2 for definitions) with the target compositions of NaAlSiO₄ and AlSiO_3.5, respectively. For some formulations, an HOAc catalyst was implemented to reduce gelation times. A series of different parameters were evaluated during the synthesis including the Na‐precursor, cosolvent alcohols (e.g., methanol, ethanol, isopropanol) utilized to suspend the Na and Al precursors, batch size, mixing times, and the mixing environment (i.e., in air or within a nitrogen glovebox). Following gelation, gels were removed from the casting vials, diced, and rinsed with 50/50 ethanol/DIW (10×), followed by solvent exchanges to pure ethanol (10×), exchanged in liquid CO₂ (10×), then the vessel was taken to the supercritical CO₂ state, and the CO₂ was vented as a gas, yielding aerogels. Several batches resulted in high SSA values ranging from 517 to 882 m² g⁻¹ for the Na‐Al‐Si‐O aerogels and up to 942 m² g⁻¹ for Al‐Si‐O aerogels. Examples of each alcogel and aerogel are shown in Figure S2 (Supporting Information).

Following gel synthesis, different approaches were utilized to load Ag into the gel network including: 1) soaking as‐made (AM) gels in AgNO₃ solutions followed by drying (Ag‐impregnation), 2) reducing the Ag⁺ in gels that underwent Ag‐impregnation to Ag⁰ under a flowing stream of 2.7%H₂/Ar gas (Ag‐reduction), and 3) thiolating gels that experienced Ag‐impregnation and Ag‐reduction (Ag‐functionalization). The appearances of these materials afterward were notably different in color (note that pictures of the materials can be found in the original work).^{[ 2 ]} During chemical analysis of the gels after Ag‐impregnation, the Na concentrations in the Na‐Al‐Si‐O base gels were really low, and nearly 0 mass% in some cases. After additional Ag‐exchange testing, the Ag⁺ was found to replace the Na⁺ in the gel network. After analyzing the Al‐Si‐O gels (those without Na), it was clear that the Ag‐content was much lower at 12 at% versus 35 at% for Ag‐functionalized gels (starting from Na‐Al‐Si‐O gels containing Na). Analyses with TEM and powder X‐ray diffraction (PXRD) on functionalized gels revealed discrete and consistently sized Ag⁰ nanoparticles present within the gels, which ranged between ≈3 and 14 nm in diameter (see Figure 5 ). Iodine loadings were conducted in a polymer autoclave containing the sample and 99.99% pure iodine crystals at 150 °C over the course of 24 h. To remove physisorbed iodine, the iodine was removed from the container, and the samples were placed back into the oven without the iodine for ≈1 h.

Figure 5.

a–e) TEM images and f–i) powder X‐ray diffraction patterns from Ag‐functionalized Na‐Al‐Si‐O aerogels and an Al‐Si‐O aerogel (AS‐1‐SH‐Ag⁰). These data show that the Ag present within these gels is crystalline Ag⁰. Adapted with permission.^{[ 2 ]} Copyright 2017, American Chemical Society.

Figure 6a provides visual appearances for the thiolated and Ag⁰‐functionalized Na‐Al‐Si‐O aerogels both inside and outside the H₂/Ar Ag‐reduction apparatus. The PXRD patterns of the as‐made, Ag‐functionalized, and iodine‐loaded Na‐Al‐Si‐O aerogels are provided in Figure 6b. The AM aerogel had a broad diffraction peak centered around 29° 2θ, which is indicative of an amorphous network without long‐range order. The Ag‐functionalized aerogel did not have the same diffraction peak as the as‐made aerogel, but rather had some broad peaks centered around 38°–46° 2θ and 65° 2θ, which corresponded well with crystalline Ag⁰ (Fm $\overline{3}$ m, ICSD No. 53761)^{[ 77 ]} with an average crystallite size of 3.2 nm. The PXRD pattern for the iodine‐loaded aerogel showed very strong diffraction peaks for AgI as verified by a match with pure AgI and contained a mixture of hexagonal (P6₃ mc, ICSD No. 62790)^{[ 78 ]} and cubic (F $\overline{4}$3m, ICSD No. 61542)^{[ 79 ]} crystalline morphologies.

Figure 6.

a) Pictures of thiolated/Ag‐impregnated and Ag‐functionalized Na‐Al‐Si‐O aerogels within the Ag‐reduction chamber (under flowing H₂/Ar); pictures of thiolated/Ag‐impregnated, Ag‐functionalized, and iodine‐loaded Na‐Al‐Si‐O aerogels. b) PXRD patterns for as‐made, Ag‐functionalized, and iodine‐loaded Na‐Al‐Si‐O aerogels; a refinement of Ag (ICSD No. 53761)^{[ 77 ]} is shown in addition to the PXRD pattern of 99.999% pure AgI. c) Iodine uptake of various samples including thiolated/Ag‐functionalized (SH‐Ag⁰), Ag‐reduced (Ag⁰), thiolated/Ag⁺‐impregnated (SH‐Ag⁺), thiolated (SH), and silver mordenite as a comparison (shown as AgZ). d) Comparison of SSA values for samples at different stages during the synthesis process including as‐made (AM), heat‐treated at 350 °C for 30 min (HT), SH, Ag⁺, Ag⁰, SH‐Ag⁺, and thiolated/Ag‐reduced (SH‐Ag⁰ or Ag‐functionalized); included are TEM micrographs showing the as‐made Na‐Al‐Si‐O and Ag‐exchanged (Ag‐Al‐Si‐O) aerogels. Adapted with permission.^{[ 2 ]} Copyright 2017, American Chemical Society.

The iodine‐loading capacities for these materials were rather high based on mass uptake and confirmed with scanning electron microscopy (SEM) and energy‐dispersive X‐ray spectroscopy (EDS) analysis showing high Ag‐utilization (I:Ag molar ratio). This means that the iodine loadings were consistent with the Ag‐loadings. The loadings based on mass uptake on a $m_{\text{I}}{m}_{\text{s}}^{-1}$ basis (mass of iodine per mass of starting sample) are shown along with the mass% of iodine measured in the sample via SEM‐EDS analysis for comparison in Figure 6c. Also included for comparison are data collected for Ag⁰Z. The iodine loadings ($m_{\text{I}}{m}_{\text{s}}^{-1}$) for Ag‐impregnated (Ag⁺), Ag‐reduced (Ag⁰), thiolated/Ag‐impregnated (—S—Ag⁺), and thiolated/Ag‐reduced (—SH + Ag⁰ or Ag‐functionalized) were 0.517, 0.555, 0.404, and 0.501 $m_{\text{I}}{m}_{\text{s}}^{-1}$, respectively. The iodine loadings were lower for thiolated versus unthiolated gels in each dataset (e.g., thiolated Ag‐impregnated vs Ag‐impregnated) and this is partly based on the added organic moieties present from the 3MPTS functionalization step (i.e., a diluent). It is also clear that the process of reducing the Ag from Ag⁺ to Ag⁰ has an added effect of increased iodine loadings from 7% to 24% for unthiolated and thiolated aerogels, respectively.

Based on these iodine loading data, extra processing steps could be avoided to yield a high‐functioning sorbent without utilizing thiolation and Ag‐reduction (Ag⁺ → Ag⁰). This expedites the sample preparation time and produces a highly functional base sorbent. However, there is also an added benefit to avoiding these steps as they decrease SSA values (Figure 6d). The most detrimental process step for lowering the SSA values is the heat‐treatment process (350 °C for 30 min in air) at the front end that is used to strengthen the gel network so that it does not collapse during the thiolation and/or Ag⁺‐impregnation steps. This heat treatment was also adopted for unthiolated gel processing. This step could be avoided but would likely result in gel damage during subsequent steps such as the Ag‐impregnation step. The second most detrimental step resulting in lowered SSA values was the thiolation process.

Following the 2017 study,^{[ 2 ]} additional comparisons were made in a 2020 study^{[ 80 ]} between Ag‐loaded aluminosilicate aerogels, heat‐treated aerogels, and xerogels. A critical difference between the 2020 study^{[ 80 ]} and the 2017 study^{[ 2 ]} is that the new work utilized a simplified process. For instance, no Ag reduction (using H₂/Ar) and no thiolation were implemented, and xerogels were introduced into the process. Based on the 2017 study,^{[ 2 ]} iodine loadings were still determined to be high when the Ag⁺ was exchanged into the base material (0.517 $m_{\text{I}}{m}_s^{-1}$) without reducing the silver to Ag⁰.

Here, similar base materials in the alcogel state (see Figure 2e) were either converted to aerogels in a critical point dryer (Figure 2g) or dried in a vacuum desiccator to yield xerogels (Figure 2f‐2). Another difference in this study was that the xerogels retained nearly the full SSA (i.e., 531 m² g⁻¹) as the aerogel counterparts (598 m² g⁻¹) due to some process improvements based on the 2017 study.^{[ 2 ]} These improvements included a longer aging time for the alcogels (several months vs a few days) and ethanol removal under vacuum within a desiccator instead of drying them in air, where drying in air is a more energetic process. After Ag⁺‐exchange, the xerogels retained higher SSA values (i.e., 240 m² g⁻¹) compared to the aerogels that were not heat treated (i.e., 120 m² g⁻¹) and even the heat‐treated aerogels (i.e., 150 m² g⁻¹). Note that the heat treatment process was performed only on some of the aerogels to improve the mechanical rigidity of the base material (as discussed previously).

The more important comparison is in the iodine uptake values for these materials. The aerogels, heat‐treated aerogels, and xerogels showed iodine uptake values of 0.405, 0.377, and 0.327 $m_{\text{I}}{m}_{\text{s}}^{-1}$, respectively. These values are not as high as those achieved in the 2017 study,^{[ 2 ]} but the base materials differed in 2020 study;^{[ 80 ]} this difference was attributed to batch‐to‐batch variability and differences in aging times used between the different sets of experiments.

The main goal of the 2020 study^{[ 80 ]} was to evaluate whether or not the iodine uptake of Ag‐loaded aerogels and xerogels differed significantly. Eliminating extra processing steps during gel synthesis drastically simplified the process and allowed for the scale‐up of sorbent production volumes. Since aerogel production requires an autoclave, this adds additional pressure hazards to the process, but also makes scale‐up more difficult and expensive than it is for xerogel production. Finally, the xerogels are quite mechanically robust, especially compared to the base aerogels and even the heat‐treated aerogels.

3.2.2. Ag‐Free Aluminosilicate Aerogels and Xerogels for Iodine Capture

A follow‐up study^{[ 47 ]} to the Ag‐loaded aluminosilicate study was conducted to evaluate non‐Ag getters in aluminosilicate aerogels and xerogels. The main driver for this study was to find alternative getters to Ag because Ag is a hazardous material controlled in the United States by the Environmental Protection Agency under the Resource Conservation and Recovery Act,^{[ 81 ]} complicating the waste form disposal process. Also, Ag is also a precious metal, making large‐scale sorbent production an expensive venture.

In this study, several different getters materials were evaluated based on thermodynamic predictions from metal‐iodide complexation ($\mathrm{\Delta }{G}_{\text{f},\text{I}}^{°}$) using Gibb's free energies compared to the metal‐oxide complexation ($\mathrm{\Delta }{G}_{\text{f},\text{O}}^{°}$) with HSC Chemistry (v9, Outotec, Finland). Calculations performed as part of that study showed that only 14 species in the database showed a more favorable metal‐iodide complexation over the metal‐oxide complexation including ($\mathrm{\Delta }{G}_{\text{f},\text{I}}^{°}$ < $\mathrm{\Delta }{G}_{\text{f},\text{O}}^{°}$) including: Cs⁺ (T _all), K⁺ (T _all), Rb⁺ (T _all), Ag⁺ (T _all), Ba²⁺ (T _all), Sr²⁺ (25 °C ≤ T ≤ 325 °C), Hg⁺ (T _all), Hg²⁺ (T _all), Tl⁺ (T ≥ 400 °C), Eu²⁺ (25 °C ≤ T ≤ 400 °C), Pb²⁺ (25 °C ≤ T ≤ 100 °C), Pd²⁺ (T _all), Pt²⁺ (25 °C ≤ T ≤ 150 °C), and Pt⁴⁺ (T = 25 °C).

Some of these species were incorporated into the sodium aluminosilicate scaffold using ion exchange processes with different solvent systems and the final selection of getters included Cs⁺, Cu²⁺, Fe³⁺, K⁺, Li⁺, Rb⁺, Sb³⁺, Sn²⁺, and Sn⁴⁺ (in addition to Ag⁺), based on promising data in the literature.^{[ 2} , ⁶ , ⁶⁸ , ⁶⁹ , ⁸² , ⁸³ , ⁸⁴ , ^{85 ]} The ion exchange process was typically depicted by the exchanged gel showing a loss of Na⁺ to the solution during the exchange as the target cations were loaded into the scaffold. The most effective materials at capturing iodine based on gravimetric uptake (measured by mass uptake) and chemical uptake (quantified by EDS analysis) were Cu²⁺ (CuSO₄), Sn²⁺ [Sn(II) acetate], and Sn⁴⁺ [i.e., colloidal SnO₂, Sn(IV) acetate] [in addition to Ag⁺ (AgNO₃)] (see Figure 7 ); note that the species in parenthesis are the salts added during ion exchange. More work is needed to fully understand the mechanisms of these exchanges for the different getter systems.

Figure 7.

Comparison of chemical uptake based on EDS data (m%_I,EDS) and gravimetric uptake (m%_I,g). The left side of the figure (shaded in gray) shows data where negative gravimetric uptake and positive chemical uptake were observed. Reproduced with permission.^{[ 47 ]} Copyright 2020, American Chemical Society.

4. Chalcogen‐Based Aerogels (Chalcogels) for Radionuclide Capture

The two primary theories to explain the high affinity of chalcogels for capturing I_2(g) and heavy metals are centered around physisorption and chemisorption. From a physisorption perspective, the selective affinity of the chalcogen‐based aerogel backbone for I_2(g) can be explained by Pearson's hard/soft acid‐base (HSAB) principle.^{[ 86} , ^{87 ]} The HSAB principle evaluates electron‐pair donating Lewis bases and electron‐pair accepting Lewis acids to predict complexation behaviors of cations and ligands.^{[ 86 ]} To compare species, the chemical hardness (η) can be calculated using the ionization energy (I) and electron affinity (A) for a given species “s” using Equation (4). If two species have similar chemical hardness values, where one is a Lewis acid and the other is a Lewis base, the HSAB principle predicts that these two species will share an affinity. Table 4 presents some different values for a range of Lewis acids and bases as well as some neutral species.^{[ 87} , ⁸⁸ , ⁸⁹ , ^{90 ]} These data suggest that the chalcogen affinity for I_2(g) would rank in the order of Te > Se > S, but to date, only S‐based chalcogels have been evaluated for this application

$${\eta }_{\text{s}}=1\text{/}2\left(I_{\text{s}}-A_{\text{s}}\right)$$ (4)

Table 4.

Tabulated values of I _s, A _s, χ_s, and η_s for various species, “s,” sorted by η_s from hard to soft. Here, χ denotes the absolute electronegativity

Acid/base	Species	I s	A s	χs	ηs	Ref(s).
Acid	K+	31.63	4.34	17.99	13.65	[ 87 , 88 ]
	Zn2+	39.7	17.96	28.8	10.8	[ 87 ]
	HCl	12.7	−3.3	4.7	8.0	[ 89 ]
	Hg2+	34.2	18.75	26.5	7.7	[ 87 ]
	U4+	45.77	31.06	38.415	7.4	[ 90 ]
	CO2(g)	13.8	0	6.9	6.9	[ 87 ]
	Cu+	20.3	7.72	14.0	6.3	[ 87 ]
	HI(g)	10.5	0.0	5.3	5.3	[ 89 ]
	HNO3	11.03	0.57	5.8	5.2	[ 89 ]
	I2(g)	9.3	2.6	6.0	3.4	[ 87 ]
Neutral	CH3I	9.5	0.2	4.9	4.7	[ 89 ]
	Cl2(g)	11.6	2.4	7.0	4.6	[ 89 ]
Base	S	10.36	2.08	6.22	4.12	[ 87 ]
	Se	9.75	2.02	5.89	3.86	[ 87 ]
	Te	9.01	1.97	5.49	3.52	[ 87 ]

The chemisorption capture route is based on evidence from the formation of metal‐iodide complexes with chalcogel constituents following the capture of I_2(g). Previous studies^{[ 6} , ⁶⁸ , ⁶⁹ , ^{82 ]} revealed Sn‐based and Sb‐based chalcogels from iodide species including SnI₄, SnI₄(S₈)₂, and SbI₃, which were easily detectable with PXRD. The formation of these iodide species allows for a better understanding of how the iodine is interacting with these types of chalcogels.

Chalcogels have very porous networks and appear very similar to oxide‐based counterparts when observed with SEM and TEM (see Figure 8 ). In some cases, crystalline diffraction has been observed with selected area diffraction (SAD), and this is presumed to be surface oxidation of the materials.

Figure 8.

(Top left) SEM and (others) TEM micrographs (with selected area diffraction patterns) for Sn₂S₃, Co_0.7Bi_0.3MoS₄, Co_0.7Cr_0.3MoS₄, Co_0.5Ni_0.5MoS₄, and PtGe₂S₅ chalcogels. Reproduced with permission.^{[ 6 ]} Copyright 2013, American Chemical Society.

4.1. Chalcogels for Iodine Capture

To date, a wide range of chalcogel chemistries have been evaluated for the capture of I_2(g) in dilute and/or concentrated streams. Compositions that have been evaluated using concentrated I_2(g) streams in a vacuum desiccator at room temperature include PtGe₂S₅,^{[ 15 ]} Sn₂S₃,^{[ 6} , ⁶⁸ , ^{69 ]} CoNiMo₂S₈,^{[ 6 ]} NiMoS₄,^{[ 82 ]} CoMoS₄,^{[ 82 ]} Sb₄Sn₃S₁₂,^{[ 82 ]} Zn₂Sn₂S₆,^{[ 82 ]} and KCoS_x.^{[ 82 ]} For consistency with other sorbents discussed herein, the $m_{\text{I}}{m}_s^{-1}$ loading capacities for iodine are provided in Table 5 for these chemistries.^{[ 15} , ⁶⁸ , ^{82 ]}

Table 5.

Summary of chalcogel chemistries evaluated for I_2(g) uptake in saturated environment showing the gel chemistry (sorted alphabetically), the sample identification (ID) from the original work, the measured SSA value for that chalcogel, the SSA _eq value [see Equation (1)], the I_2(g) loading based on mass uptake within the vacuum desiccator, and the corresponding reference for the original work

Chalcogel	Sample ID	SSA [m2 g−1]	SSA eq [m2 g−1]	I2(g) loading [$m_{\text{I}}{m}_s^{-1}$]	Ref.
CoMoS4	CoMoS	360	848	2.00	[ 82 ]
KCoSx	KCoS	350	–	1.60	[ 82 ]
NiMoS4	NiMoS	490	1154	2.25	[ 82 ]
PtGe2S5	Cg‐5C	360	1200	2.39	[ 15 ]
Sb4Sn3S12	SbSnS	240	817	2.00	[ 82 ]
Sn2S3	SnSg+I	270	999	2.13	[ 68 ]
Zn2Sn2S6	ZnSnS	400	1244	2.25	[ 82 ]

Additionally, iodine capture was evaluated using a dilute I_2(g) stream environment of 4.2 ppm (by volume) in air as the carrier gas. These experiments were conducted using a DYNACAL iodine permeation tube with a permeation rate of 22.8597 ng s⁻¹ at 100 °C. This stream was connected to a separate furnace at 150 °C through a heated transfer line where chalcogel sorbents resided; the vent line for this system was placed in 0.1 m NaOH solution that was replaced periodically; this setup can be seen in Figure 9a. Iodine capture efficiency was determined by comparing the iodine breakthrough into the NaOH solution as a function of time, with the sorbent in the column, and those values were compared to the iodine concentration when the sample was not present. Solutions were analyzed using inductively coupled plasma mass spectrometry (ICP‐MS) to determine the iodine concentration. The compositions evaluated using this setup include PtGe₂S₅,^{[ 6} , ^{15 ]} CoNiMo₂S₈ (see Figure 9c),^{[ 6 ]} and Sn₂S₃ (see Figure 9d).^{[ 6 ]} In all cases, the uptake over the test duration was > 99 mass% with Sn₂S₃ showing the highest uptake over the full test duration demonstrated at >99.5% and increasing with time (see Figure 9b). In a separate study by Subrahmanyam et al.^{[ 91 ]} (NH₄)₂MoS₄ chalcogels were also demonstrated to have a high affinity for iodine vapors.

Figure 9.

a) Experimental setup for 4.2 ppm iodine capture experiments using DYNACAL permeation tube, b) a summary of the capture as a function of time for PtGe₂S₅, CoNiMo₂S₈, and Sn₂S₃,^{[ 6} , ^{15 ]} c) a close‐up picture of the CoNiMo₂S₈ chalcogel after the uptake test, and d) a close‐up picture of the PtGe₂S₅ chalcogel after the uptake test; Adapted with permission.^{[ 6 ]} Copyright 2013, American Chemical Society. e) Schematic showing chalcogel‐PAN composite synthesis process and f) iodine loading as a function of time for two composites and the pure PAN (no chalcogel added); Reproduced with permission.^{[ 68 ]} Copyright 2014, American Chemical Society.

One way to increase the mechanical integrity of a sorbent is to embed the active sorbent in a passive matrix (e.g., a polymer‐based scaffold) to create a composite material. Here, the passive matrix essentially dilutes the capacity of the active sorbent on a per‐mass basis since a part of the composite is made up of a passive matrix that either does not interact with the contaminant(s) of interest or only interacts slightly. In a previous study,^{[ 68 ]} this approach was used with Sn₂S₃ chalcogel powders to make chalcogel‐polyacrylonitrile (PAN) composites for iodine capture (Figure 9e). Some benefits from this approach are that the PAN matrix is very porous, stable under elevated temperatures, mechanically robust, and very lightweight, adding minimal mass to the active chalcogel sorbent. In this study, Sn₂S₃‐PAN composites were made containing 30, 50, and 70 mass% chalcogel in the PAN. The iodine capacity scaled quite linearly with the amount of Sn₂S₃ present in the composite,^{[ 92 ]} and the loading rates changed with different Sn₂S₃ loadings (Figure 9f). These results demonstrate both the passive nature of the PAN matrix and the need for PAN minimization to maximize the iodine capacity in the final composite sorbent.

4.2. Chalcogels for the Capture of Uranium and Technetium

Studies were done to evaluate the uptake of uranium‐238 (i.e., ²³⁸UO₂ ²⁺) and technetium‐99 (i.e., ⁹⁹TcO₄ ⁻) ions from solution using the following chalcogel chemistries: Co_0.7Bi_0.3MoS₄ (CoBiMoS), Co_0.7Cr_0.3MoS₄ (CoCrMoS), Co_0.5Ni_0.5MoS₄ (CoNiMoS), PtGe₂S₅ (PtGeS), and Sn₂S₃ (SnS).^{[ 6 ]} For these experiments, ≈0.05–0.1 g of each chalcogel were loaded into separate 20 mL glass vials with 9 mL DIW; a blank vial was also loaded with 9 mL of DIW. This was followed by the addition of 1 mL solution containing either 10⁻⁵ m ⁹⁹TcO₄ ⁻ added as NaTcO₄ or 10⁻⁵ m ²³⁸UO₂ ²⁺ added as UO₂(NO₃)₂ for final concentrations of 10⁻⁶ m for each target radionuclide. Sample vials were sealed and mixed end‐over‐end for 7 d. Then the supernatant was collected through a syringe with a 0.45 µm filter, and the analyte concentrations were analyzed with ICP‐MS. Capture efficiencies and K _d values were measured using Equations (3) and (5) (see Section 3), respectively, where m is the mass of sorbent material (V/m ≈ 1000 mL g⁻¹) and other variables were defined in Section 3.

$$\text{Efficiency}=100\ast \frac{C_{\text{i}}-C_{\text{t}}}{C_{\text{i}}}$$ (5)

The efficiency data and K _d values from these tests are shown in Table 6 . The values revealed a wide range of values for the different chalcogel chemistries where CoBiMoS, PtGeS, and SnS worked well for both ⁹⁹TcO₄ ⁻ and ²³⁸UO₂ ²⁺ at ≥87.3% fractional uptake and K _d values of ≥1.5 × 10³. The highest values were observed for PtGeS for both species with K _ds of 3.6 × 10⁴ for ⁹⁹TcO₄ ⁻ and 9.4 × 10⁴ for ²³⁸UO₂ ²⁺ along with efficiencies of 98.0% and 99.4%, respectively. The lowest uptake values were found with CoCrMoS at 57.3% and 68.1% efficiencies for ⁹⁹TcO₄ ⁻ and ²³⁸UO₂ ²⁺, respectively. These data provide a demonstration that chalcogels could prove useful for the remediation of these radionuclides from aqueous solutions.

Table 6.

Summary of solution uptake of ⁹⁹TcO₄ ⁻ and ²³⁸UO₂ ²⁺ with chalcogels including the efficiency (Eff. %); [see Equation (3)] and the K _d values [see Equation (4)]. Reproduced with permission.^{[ 6 ]} Copyright 2013, American Chemical Society

Sample ID	99Tc uptake				238U uptake
	Eff. [%]	K d [mL g−1]	Final pH	Color	Eff. [%]	K d [mL g−1]	Final pH	Color
CoBiMoS	94.0	1.68 × 103	2.31	Very blue	94.9	1.81 × 103	2.26	Faintly blue
CoCrMoS	57.3	2.88 × 102	2.78	Faintly pink	68.1	3.15 × 102	2.72	Barely pinkish
CoNiMoS	62.2	1.62 × 102	4.12	Pale blue	88.2	8.08 × 102	9.06	Pale blue
PtGeS	98.0	3.61 × 104	4.80	Faintly orange	99.4	9.43 × 104	4.44	Faintly orange
SnS	87.3	1.49 × 103	5.47	Faint yellow	99.1	2.31 × 104	5.53	Faintly yellow
Blank	–	–	5.59	Clear	–	–	5.18	Clear

4.3. Chalcogels for the Capture of Heavy Metals

Recent work showed promise for chalcogels for heavy metal remediation; many of these heavy metals could be considered as byproducts to radiological processes and, thus are included in the current review. In a study by Subrahmanyam et al.^{[ 91 ]} (NH₄)₂MoS₄ chalcogels were demonstrated to have a high affinity for Hg vapors, forming crystalline HgS upon capture with a final composition of Hg_1.9MoS₄ after complete exchange of the NH₄ with Hg. In a study by Pala and Brock,^{[ 93 ]} ZnS chalcogels were used to remove Pb²⁺ and Hg²⁺ from aqueous solutions. Oh et al.^{[ 59 ]} showed that Zn₂Sn_xS_{2 x +2} (x = 1, 2, 4) chalcogels can be used to remove Hg²⁺, Pb²⁺, Cd²⁺, Cu²⁺, Zn²⁺, and Fe²⁺ from aqueous solutions with very high K _d values ranging from 7 × 10³ (Fe²⁺) to 2.8 × 10⁷ (Hg²⁺).

4.4. Chalcogels for the Capture of Noble Gases

Noble gases are byproducts of nuclear fission, namely, Xe and Kr. Under 40 CFR 190,^{[ 94 ]} the U.S. Environmental Protection Agency requires the capture ⁸⁵Kr evolved from a nuclear fuel reprocessing facility, if one were to be built in the United States. The concentration of Xe present within irradiated nuclear fuel is nominally 10× the total Kr concentration and this combined with the unreactive nature of noble bases makes the selective capture of Kr even more challenging. The traditional option for capturing noble gases has been cryogenic distillation,^{[ 95} , ⁹⁶ , ⁹⁷ , ⁹⁸ , ^{99 ]} but new MOF sorbents^{[ 100} , ¹⁰¹ , ^{102 ]} have shown promise as alternatives to distillation.

A paper by Subrahmanyam et al.^{[ 103 ]} revealed that chalcogels could be used to capture Xe from the noble gas mixture. Here, four different chalcogels were utilized including (NH₄)_0.03MoS₄ (MoS_x), Na_0.3Sb₂S₃ (SbS‐I), K_0.15Na_0.3Sb₂S_2.5 (SbS‐II), and Na_0.1Sb₂S₃ (SbS‐III) with SSA values of 128–290 m² g⁻¹. The MoS_x chalcogel showed the highest uptake for both Xe and Kr as shown in Figure 10a,b, respectively. Calculations were conducted using the ideal adsorbed solution theory (IAST) to look at Xe/Kr selectivities among the four chalcogels and the MoS_x chalcogel exhibited the highest value of 6.0 compared to selectivities of 2.0–2.8 for all Sb‐S chalcogels. It is hypothesized that the differences in selectivities between the Mo‐S and Sb‐S chalcogels are possibly due to the differences in metal ion oxidation state (i.e., Mo⁴⁺ vs Sb³⁺) as well as crystal radii.

Figure 10.

a) Xe and b) Kr adsorption isotherms recorded at 273 K up to 1 bar. Adapted with permission.^{[ 103 ]} Copyright 2017, American Chemical Society.

5. Heavy Metal Capture with Carbon‐Based Aerogel Composites

For removal of heavy radionuclides from aqueous solutions, different carbon‐based functionalized aerogel composites have been developed to improve adsorption capacities while maintaining high porosities and high SSA values while also allowing for the incorporation of various getters in the structures.^{[ 104} , ¹⁰⁵ , ¹⁰⁶ , ¹⁰⁷ , ¹⁰⁸ , ¹⁰⁹ , ^{110 ]} Table 7 shows the adsorption capacities of radionuclides by various sorbents in the literature;^{[ 105} , ¹⁰⁶ , ¹⁰⁷ , ^{108 ]} also included in Table 7 are additional data for similar compounds without aerogels included for comparison.^{[ 111} , ^{112 ]}

Table 7.

Comparison of the adsorption capacities of various sorbents on radioactive and surrogate ions

Type	Sorbents	Target species	Species capacity [mg g−1]	Solution pH	Temperature [K]	Ref.
With aerogels	Graphene aerogel	U	131	5.5	298	[ 105 ]
	PANI‐graphene aerogel	U	242	5.5	298	[ 105 ]
	Fe0‐PANI‐graphene aerogel	U	350	5.5	298	[ 105 ]
	Fungal hypha‐graphene oxide aerogel	U	288	5	293	[ 106 ]
	Konjac glucomannan/graphene oxide aerogel	U	513	5	313	[ 107 ]
	SiO2‐ZrO2‐calcium alginate aerogels	Ce	232	7	308	[ 108 ]
	Graphene oxide/polypyrrole	U	147	5	298	[ 117 ]
	Na2Ti3O7‐carbon aerogel	Rb	94.9	5.5	298	[ 110 ]
	Na2Ti3O7‐carbon aerogel	Cs	193	5.5	298	[ 110 ]
Without aerogels	Carbonaceous nanofiber	U	125	4.5	298	[ 111 ]
	Carbonaceous nanofiber	Eu	91	4.5	298	[ 111 ]
	Graphene oxide	U	56	3.8	298	[ 112 ]
	Graphene oxide/α‐MnO2	U	185	3.8	298	[ 112 ]
	Graphene oxide	Th	154	3.8	298	[ 112 ]
	Graphene oxide/α‐MnO2	Th	498	3.8	298	[ 112 ]

Coleman et al.^{[ 104 ]} developed a granulated activated carbon (GAC)‐silica aerogel composite for uranium removal from aqueous solution. Hydrophobic GAC‐aerogel composites were synthesized and modified with phosphoric acid, phosphonate, or Ca(NO₃)₂, and their uranium adsorption capacities were evaluated in 100 ppb uranium and pH 7 surrogate solution.^{[ 104 ]} The results showed that GAC‐aerogel composites were better adsorbents compared to normal GAC.

Chen et al.^{[ 105 ]} developed a Fe⁰‐polyaniline‐graphene aerogel composite (Fe⁰‐PANI‐GA) for U⁶⁺ removal from acidic solutions. Graphene aerogel is composed of cross‐linked graphene sheets with robust porous structures that can incorporate getters and has been studied for removal of heavy metal ions.^{[ 113 ]} Zero‐valent iron (Fe⁰) is an effective adsorbent for heavy metal ions due to more reactive sites on the surface,^{[ 114} , ^{115 ]} and polyaniline contains abundant primary and secondary amino groups that help assist the adsorption of heavy metal ions.^{[ 116 ]} The Fe⁰‐PANI‐GA composite was more effective for removal of U⁶⁺ compared to GA or PANI‐GA alone in acidic solution, and the maximum sorption capacity was 350 mg g⁻¹ in pH 5 solution at room temperature (see Table 7).^{[ 105 ]}

Liu et al.^{[ 110 ]} developed Na₂Ti₃O₇‐carbon aerogel for removal of Rb⁺ and Cs⁺ cations from aqueous solution. The adsorption mechanism involves replacement of Na⁺ with Rb⁺ and Cs⁺ cations in Na₂Ti₃O₇ and complexation of hydroxyl functional groups from Na₂Ti₃O₇‐carbon aerogel with Rb⁺ and Cs⁺ cations.^{[ 110 ]} The maximum adsorption capacity of Na₂Ti₃O₇‐carbon aerogel for Rb⁺ and Cs⁺ cations were 1.11 mmol g⁻¹ (94.9 mg g⁻¹) and 1.45 mmol g⁻¹ (193 mg g⁻¹), respectively, from a solution of pH 5.5.^{[ 110 ]}

Li et al.^{[ 106 ]} used a combination of microorganisms and aerogels to develop an adsorbent of fungal hypha‐graphene oxide aerogel composite for uranium removal from aqueous solutions. Fungal hypha contain a large amount of phosphonate, hydroxyl, and amine groups on the cell wall that have strong chelation with uranium ions.^{[ 106 ]} The sorption capacity of fungal hypha‐graphene oxide aerogel composites for U⁶⁺ ions in a solution of pH 5 was 288 mg g⁻¹.^{[ 106 ]}

Chen et al.^{[ 107 ]} developed konjac glucomannan/graphene oxide composite aerogel for U⁶⁺ removal from simulated wastewater. Konjac glucomannan is a renewable natural resource that can adsorb radionuclides, but with a relatively low adsorption capacity. By combining with graphene oxide and aerogel, its absorption capacity can be increased significantly.^{[ 107 ]} The maximum adsorption capacity of U⁶⁺ from pH 5 solution at 313 K was 513 mg g⁻¹.^{[ 107 ]}

Zhao et al.^{[ 108 ]} removed cerium, a surrogate of plutonium and minor actinides, from an aqueous solution using SiO₂‐ZrO₂‐calcium alginate aerogels and showed that Ca and Zr sites were occupied by Ce⁴⁺ cations with the maximum adsorption capacity of 232 mg g⁻¹. Calcium alginate is a natural biological polymer that can be extracted from brown algae.

Zhu et al.^{[ 109 ]} demonstrated that low cost and non‐toxic cotton‐carbon aerogel with high cellulose content can be used as an adsorbent for removal of Sr²⁺ cations from pH 5–7 solutions. The maximum removal efficiency of cotton‐carbon aerogel for Sr²⁺ removal was 60.2% from a solution of pH 6, and cotton‐carbon aerogel was reusable for five adsorption and desorption cycles.^{[ 109 ]}

6. Summary and Conclusions

This progress report provides a review of recent work in the area of functional aerogels and xerogels for environmental remediation applications. The materials discussed herein include sorbents for liquid and/or gaseous streams. Sorbents discussed include metal‐functionalized silica‐based aerogels and xerogels, sulfide‐based aerogels, and carbon‐containing aerogel composites for target species such as Cd²⁺, Ce⁴⁺, Cs⁺, Cu²⁺, Fe²⁺, Hg²⁺, Hg_(g), I⁻, IO₃ ⁻, Kr, Pb²⁺, Rb⁺, Sr²⁺, ⁹⁹Tc⁷⁺, U⁶⁺, Xe, and Zn²⁺. These sorbents have a range of specific surface areas, porosities, and capacities for these contaminants.

Depending on the application, different materials are more suitable than others. For instance, some aerogels are quite fragile and friable; these might not be ideal for applications requiring high‐flow gas streams or extreme‐pH aqueous environments. For such environments, additional techniques can be implemented such as embedding the active sorbent in a passive matrix, as was done with the chalcogel‐PAN composites; this added mass of the inactive matrix will decrease the capacity of the overall composite, but with the added benefit of increased mechanical integrity.

Overall, oxide‐based aerogels and xerogels are likely less expensive to make than chalcogels, due to the commercially availability of a wide range of alkoxide precursors and a very limited availability of chalcogel precursors; additionally, commercially available chalcogel precursors also tend to be rather expensive. The added benefit of chalcogels, however, is that they are oftentimes inherently “functionalized” for these types of applications where the as‐made sorbents can be used to capture species of interest. In contrast, oxide‐based gels often require the addition of a getter species, e.g., Ag⁺/Ag⁰.

The primary conclusion of this work is that plenty of open spaces exist on this topic in the literature and new applications are being discovered regularly. Additional topics of interest and challenges in this area are plentiful, some of which are described below. For instance, future work should focus on finding non‐hazardous metal getters. Silver is a hazardous material as well as a precious metal. With recent work showing promise for other getters such as Cu, Sn, and Sb, this provides an opportunity for cheaper and less hazardous sorbents. A limitation with aerogel production is the requirement for critical point drying using autoclaves and supercritical fluids. This can be avoided if xerogels are pursued instead. Plus, xerogels can be made with minimal SSA loss, as was discussed herein, and they tend to have significantly higher mechanical stability than the aerogel counterparts. A large‐scale production process for making xerogels would greatly improve the workflow.

Very little work has been done looking at gel‐polymer composites for these applications and this type of approach could be used to further push the limits of environmental remediation applications, e.g., under conditions where gels (without added support) might not fare well from a mechanical integrity perspective. The added disadvantage of the polymer being part of the composite is that the sorbent cannot be easily consolidated into a waste form if the polymer remains, unless it is selectively removed (e.g., by burning it off) beforehand. However, the composite sorbent could be loaded into a secondary matrix for disposal, such as a cement or grout.

This leads into the final point, which is that the fate of the contaminant‐loaded sorbent need be considered. Very little has been done to attempt consolidation of contaminant‐loaded gels into a waste form suitable for long‐term disposal. This remains an area that is wide open for future research. Since these sorbents are porous, at a minimum, they could be hot pressed into a dense monolithic waste form. However, it would be preferred if the contaminants could be removed and the base sorbent recycled for future use.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Biographies

Brian J. Riley is a senior materials scientist at PNNL, where he is also the technical team leader for the Waste Form Development Team in the Nuclear Sciences Division. His research focuses on waste form development and sorbent development. His expertise is grounded in materials design and fabrication for environmental remediation applications.

Saehwa Chong is a materials scientist working as a postdoctoral research associate at PNNL. His research focuses on waste form development. He is interested in developing engineered sorbent composites using various metals with porous minerals and gels for capturing radionuclides and toxic elements. He also works on the synthesis and characterization of various crystals that are found in nuclear waste forms.

Riley B. J., Chong S., Environmental Remediation with Functional Aerogels and Xerogels. Global Challenges 2020, 4, 2000013 10.1002/gch2.202000013