Adsorption behavior of one-dimensional coordination polymers [Cu(bipy)X]_n (X = 2Cl⁻ and SO₄²⁻) toward acid orange 7 and methyl orange

Abstract

A kind of one-dimensional coordination polymer, [Cu(bipy)X]_n (X = 2Cl⁻ or SO₄²⁻), was prepared by Cu²⁺ and 4,4′-bipyridine (bipy) through a facile method, and used as the adsorbent for removal of acid orange 7 (AO7) and methyl orange (MO) from water. Two kinds of anions, chloride ion and sulfate ion, were tested as the counter ion in preparing the [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n to balance the positive charge of [Cu(bipy)]_n²⁺, to research the impact of anions on the adsorption performance of the [Cu(bipy)]_n²⁺. Compared to [Cu(bipy)Cl₂]_n, [Cu(bipy)(SO₄)]_n can remove the dyes better both from the view of the adsorption capacity and adsorption kinetic constant. The MO can be more easily adsorbed than AO7 by both adsorbents. The adsorption capacity of [Cu(bipy)Cl₂]_n for AO7 and MO reached 848 mg g⁻¹ and 1084 mg g⁻¹, respectively. However, the adsorption capacity of [Cu(bipy)(SO₄)]_n for AO7 and MO reached 3308 mg g⁻¹ and 1521 mg g⁻¹, respectively. The dominant adsorption mechanism of the two adsorbent in removing dyes was found to be ionic exchange and the ultra high adsorption capacity of [Cu(bipy)(SO₄)]_n for AO7 was attributed to the two-step adsorption process which can be observed from the adsorption isotherm. The mechanism of the second adsorption step was speculated to be hydrophobic adsorption and π–π interaction.

---

1. Introduction

Coordination polymer (CP) are a class of organic–inorganic hybrid materials, which are mainly self-assembled by coordination of organic ligands containing oxygen or nitrogen and metal ions or metal-containing clusters.¹ By changing the ligands and metal ions, CP materials with different structures can be synthesized. Some of them are porous materials with three-dimensional skeleton structure, which usually have large specific surface area.² They are called metal organic frameworks (MOFs). Some of them have little pores with one-dimensional or two-dimensional molecular structure.³ In the area of adsorption, especially in hydrogen storage and gas removal, MOFs have attracted great interest because of their ultra high specific surface area and tunable skeleton structure.⁴ In recent years, many three-dimensional MOFs which are stable in liquid phase were used in aqueous medium as the adsorbent to remove hazardous pollutants including dyes, PPCPs and benzenes, etc. Electrostatic interactions, hydrogen bonding, π–π interactions, and other mechanisms were found in the adsorption processes.⁵ Some of them with lower surface area could remove more pollutant through electrostatic interaction. For example, in removing methyl orange (MO), MOF-235 (with surface area of 974 m² g⁻¹) has a adsorption capacity of 477 mg g⁻¹, while ethylenediamine-grafted MIL-101 (with surface area of 3491 m² g⁻¹) has a adsorption capacity of 160 mg g⁻¹.^6–8 These suggest that the effective adsorption site is another important factor of an adsorbent besides surface area.

One-dimensional CP has been seldom studied as adsorbent because they have no pores and thus little surface area. However, one-dimensional CP that was coordinated by neutral ligand and metal centre was supposed to have large amounts of adsorption sites with electrostatic force. This study choose 4,4′-bipyridine (bipy) as the neutral organic ligand, and Cu²⁺ as the coordination metal center to form [Cu(bipy)]_n²ⁿ⁺. Chloride ion and sulfate ion were selected as the anions to neutralize the positive charge of Cu²⁺, to study the impact of different anions on the adsorption performance of the CP.

Acid orange 7 (AO7) and methyl orange (MO) are two typical dyes which have been studied frequently in recent years as pollutants, as the increasing discharge of dye wastewater.^9,10 And it is well-known that dye wastewater is carcinogenic, toxic and difficult to be degraded by bio-methods.¹¹ Besides, AO7 and MO are two monovalent dyes with comparatively simple structure. It is propitious to evaluate the adsorption performance of the CP and to study the adsorption mechanisms in the adsorption processes. The molecular structures of AO7 and MO were illustrated in Fig. S1.†

2. Experimental

2.1. Materials

CuCl₂·2H₂O (99.0%) and CuSO₄ (99.0%) were purchased from Tianjin Fuchen Chemical Reagents Factory, ethanol (99.7%) was obtained from Sinopharm Chemical Reagent Co., 4,4′-bipyridine (98%) and the two dyes were purchased from Aladdin Chemistry Co. All chemicals were used without further purification.

2.2. Synthesis of [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n

The powder samples of [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n were prepared by mixing CuCl₂·2H₂O or CuSO₄ with 4,4′-bipyridine at a mole ratio of 1 : 1 in ethanol–water solution, and then being agitated for 1 h at 25 °C. The resulting solids were collected by filtration and washed repeatedly with deionized water, oven-dried at 100 °C for 24 h, then powdered and finally screened by a 100-mesh sieve. The powders were stored and used as adsorbents.

The single crystal sample of [Cu(bipy)(H₂O)₃(SO₄)]_n·2nH₂O was obtained by a hydrothermal method. The mixture of 0.1596 g CuSO₄, 0.2124 g 4,4′-bipy, 15 mL deionized water and 5 mL ethanol were placed in a 100 mL Teflon-lined stainless steel reactor and then heated in an oven at 110 °C for 24 h. The light blue needle-like crystals were obtained and a single crystal with dimensions 0.12 × 0.1 × 0.2 mm was used for data collection of X-ray diffraction.

2.3. Characterization and analysis

The BET specific surface area of [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n powders were obtained by a multipoint BET method from nitrogen adsorption isotherms determined with Automated Gas Sorption Analyzer at 77 K (Quantachrome Instruments, USA). The size distribution of the particles was determined by Mastersizer 3000 (Malvern). The ζ-potential of the samples was analyzed with a zeta potential analyzer (Zetasizer Nano ZS, Malvern). Crystal and molecular structure data of [Cu(bipy)(H₂O)₃(SO₄)]_n·2nH₂O were collected on a Rigaku R-AXIS SPIDER IP diffractometer with Mo Kα radiation (λ = 0.71073 Å). Crystal data for [Cu(bipy)(H₂O)₃(SO₄)]_n·2nH₂O:C₁₀H₁₈CuN₂O₉S, fw = 405.86, hexagonal, space group P6₅(170), Z = 6, a = 11.215 (5) Å, b = 11.215(5) Å, c = 21.4977(12) Å, α = β = 90°, γ = 120°, V = 2341.8(2) ų, D_c = 1.727 g cm⁻³; 3381 unique reflections were measured and used in refinement. The structure was solved by the direct method and refined anisotropically on F² by full-matrix least-squares techniques using the SHELXTL 97 program. All hydrogen atoms were generated geometrically. The final refinement gave R = 0.0603, R′ = 0.1385. The final difference map had peaks between −0.571 and 0.660 ų. CCDC reference 988295 contain the crystallographic data of [Cu(bipy)(H₂O)₃(SO₄)]_n·2nH₂O.

2.4. Adsorption experiments

The batch adsorption was adopted to measure the adsorption equilibrium isotherms of the adsorbents for AO7 and MO. A series of dye solutions with initial concentrations in the range of 100–2000 mg L⁻¹ were prepared. 20 mg of adsorbents were added to a 50 mL conical beaker containing 40 mL of dye solution. The pH of the samples were not adjusted artificially, and these beakers were sealed to avoid the potential water evaporation and then agitated in a thermostatic agitator bath at 300 K with a agitate speed of 200 rpm until the adsorption process reached equilibrium. Then the samples were centrifuged with a speed of 3000 rpm for 5 min and analyzed with UV-spectrophotometer at 484 and 463 nm for AO7 and MO, respectively. The concentration of sulfate ion and chloride ion in the samples was determined with the ion chromatograph (882 Compact IC plus, Metrohm, Swiss).

The adsorption kinetic study was conducted as the same to the isotherms experiments except that the initial dye concentration was fixed at 400 mg L⁻¹. And the samples with different adsorption time were centrifuged and analyzed as mentioned above.

In the combining adsorption of AO7 and malachite green (MG), mixture of 200 mg L⁻¹ of MG and a series concentration (100–800 mg L⁻¹) of AO7 were adsorbed by [Cu(bipy)(SO₄)₂]_n. After reaching equilibrium, the concentration of AO7 and MG were analyzed using a UV-spectrophotometer at 484 and 617 nm, respectively.

3. Results and discussion

3.1. Characterization of adsorbents

The prepared [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n using as the adsorbents for removal of dyes from water are blue powders with an average particle size of 5.2 μm and 6.7 μm, respectively. The BET surface area of the two materials is 5.8 m² g⁻¹ and 5.3 m² g⁻¹, respectively, which indicate the adsorbents are not porous materials as traditional adsorbents like activated carbon. Fig. 1 shows the picture of the two adsorbents and molecular structures from Single Crystal X-ray Diffraction data. In both materials, bipy acts as a bridging ligand, forming [Cu(bipy)]_n²ⁿ⁺ with the coordination centre of Cu²⁺. Their difference is the negative ions which balance the positive charges of Cu²⁺ ion. For [Cu(bipy)(SO₄)]_n, a sulfate group bond with a Cu²⁺ ion to form a Cu–O (SO₄²⁻) bond with a length of 2.650 Å. While for [Cu(bipy)Cl₂]_n, as reported by N. Masciocchi et al., a Cl⁻ ion bonds with two Cu atoms in two adjacent [Cu(bipy)]_n²ⁿ⁺ chains, forming a Cu–Cl⋯Cu bridging bond with a shorter length of 2.370 Å and a longer length of 2.884 Å.³

Fig. 1 The physical character and molecular structure of [Cu(bipy)Cl₂]_n (ref. 3) and [Cu(bipy)(SO₄)]_n.

3.2. Adsorption isotherms study

Fig. 2 shows the adsorption isotherms of AO7 and MO on the adsorbents. As it shows, both the two adsorbents exhibited high adsorption capacity on the AO7 and MO. The experimental adsorption capacity of [Cu(bipy)Cl₂]_n is 848 mg g⁻¹ and 1084 mg g⁻¹ for AO7 and MO, respectively. And the [Cu(bipy)(SO₄)]_n showed a capacity of 3308 mg g⁻¹ and 1521 mg g⁻¹ for AO7 and MO. To the best of our knowledge, these values are comparatively higher than others reported. Up to now, many adsorbents have been used for removing AO7 and MO, some of which have extremely high surface area. These previously reported values of some outstanding adsorbents are listed in Table 1.^{6–8,12–18}

Fig. 2 Adsorption isotherms of AO7 and MO on (a) [Cu(bipy)Cl₂]_n and (b) [Cu(bipy)(SO₄)]_n.

Table 1 Comparison of maximum adsorption capacities of AO7 and MO on various adsorbents

	Adsorbents	Surface area (m^2 g^−1)	Qm (mg g^−1)	References
AO7	Carbon–alumina core–shell spheres	182	210	12
	Kapok fiber oriented polyaniline	21.8	188.7	13
	Dithiocarbamate-modified starch	—	100	14
	Modified ordered mesoporous carbon	1838	569	15
	Amberlite IRA-67	—	1211.3	16
	[Cu(bipy)Cl2]n	5.8	848	This work
	[Cu(bipy)(SO4)]n	5.3	3308	This work
MO	Modified straw	—	327.9	17
	MOF-235	974	477	6 and 8
	Ethylenediamine-grafted MIL-101	3491	160	7
	Carbon nanotube-activated	534.6	149	18
	[Cu(bipy)Cl2]n	5.8	1084	This work
	[Cu(bipy)(SO4)]n	5.3	1521	This work

---

In order to further understand the adsorption behaviors of the dyes on the adsorbents. Two classical isotherm models, Langmuir¹⁹ and Freundlich,²⁰ were used to fit the adsorption isotherms.

Langmuir isotherm is given as:

and its linear form:where C_e is the concentration (mg L⁻¹) of adsorbate in liquid phase at equilibrium. Q_e is the adsorption amount of adsorbent (mg g⁻¹) at equilibrium. Q_m is the monolayer capacity of adsorbent (mg g⁻¹) and b is the Langmuir adsorption constant (L mg⁻¹).

The Freundlich isotherm is developed on the hypothesis of heterogeneous adsorption and multilayer adsorption.

The Freundlich model is formulated as:

Q_e = kC_e^1/n (3)

and its linear form:where C_e is the concentration (mg L⁻¹) of adsorbate in liquid phase at equilibrium. Q_e is the adsorption amount of adsorbent (mg g⁻¹) at equilibrium. k and n, being indicative of the extent of the adsorption and the degree of nonlinearity between solution concentration and adsorption, are Freundlich adsorption isotherm constants.

The fitted curves are shown in Fig. 2. The parameters of the two models are listed in Table 2. It can be easily figured out, that the Langmuir isotherm model fits the experimental data better than the Freundlich model from the curves and the correlation coefficients. This may indicate the adsorption happened on the adsorbents are monolayer adsorption. For Freundlich model, it can tell us a lot of information. For both adsorbents, the k (Freundlich isotherm constants) of the MO is higher than that of AO7, which indicate, compare to AO7, MO can be adsorbed by the adsorbents easier. This result can also be obtained from the isotherms in Fig. 2, because the C_e of MO is lower than that of AO7 at an equal value of Q_e.

Table 2 Parameters of the Langmuir and Freundlich models fitted for adsorption isotherms

Parameters	Langmuir				Freundlich
	Qm-exp mg g^−1	Qm mg g^−1	b L mg^−1	R^2	k (L mg^−1)^1/n (mg g^−1)	n	R^2
[Cu(bipy)Cl2]n
AO7	848	921	0.0182	0.9823	103	2.80	0.9752
MO	1084	1233	0.0240	0.9901	136	2.62	0.9341
[Cu(bipy)(SO4)]n
AO7	1318	1600	0.0264	0.9434	128	2.10	0.9385
MO	1521	1514	0.0816	0.9091	432	4.52	0.8758

---

Fig. 2 also show an interesting phenomenon that the isotherm of the AO7 adsorbed by [Cu(bipy)(SO₄)]_n showed a two-stepwise adsorption curve. It can be observed from Fig. 2b, that the second adsorption step of AO7 almost started when the adsorption amount reached the stage of the isotherm of MO. As previously discussed, MO can be adsorbed better by [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n. However, the maximal adsorption amount of AO7 by [Cu(bipy)(SO₄)]_n reached 3308 mg g⁻¹ because of the second adsorption step. As the same kind of adsorbent, [Cu(bipy)Cl₂]_n does not show such two-stepwise adsorption as Fig. 2a shows. Its adsorption capacity for AO7 is much lower than that of [Cu(bipy)(SO₄)]_n. To answer these questions, the adsorption mechanisms need to be further explored.

3.3. Mechanism discussion

AO7 and MO are anionic dyes. In our study, we also tested a cationic dye of Malachite Green (MG) as adsorbate. Although the two adsorbents can effectively remove AO7 and MO, the MG could hardly be adsorbed. These phenomena help us to figure out the adsorption mechanism favoringly. As mentioned before, the BET surface areas of both two adsorbents are small. Thus theoretically, if the adsorbents are materials with rigid structure, they could not have such high adsorption capacities for AO7 and MO. Based on the character of one-dimensional molecule structure of the two adsorbents as Fig. 1 shows, it is rational to speculate that the one-dimensional chains would depart from each other under the impact of the water and adsorbate. For such high adsorption capacities, enough adsorption sites are needed even the specific surface area is not the constraint. Considering on their coordination structures, the Cu atoms should be the most probable adsorption sites. And the interaction between adsorbate and adsorbent may be electrostatic effect after anions exchanged (SO₄²⁻, Cl⁻).

Fig. 3 shows the adsorption capacities of the two adsorbents for AO7 and MO, and the amount of corresponding exchanged anions. It shows that the adsorption capacities of [Cu(bipy)Cl₂]_n for both AO7 and MO are almost equal to the Cl⁻, while that of [Cu(bipy)(SO₄)]_n for AO7 and MO are more than SO₄²⁻. As mentioned before, AO7 and MO are monovalent dyes. Thus it can be easily explained, because Cl⁻ is a monovalent anion and SO₄²⁻ is a divalent anion. Theoretically, one mol of adsorbed adsorbate can exchange one mol of Cl⁻ or 0.5 mol of SO₄²⁻. These experiments can illustrate that the major mechanism of the adsorption is ion exchange.

Fig. 3 The experimental adsorption capacity of [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n for AO7 and MO, and corresponding exchanged anions.

Besides the ion exchange mechanism, two important phenomena are also revealed in Fig. 3. First, the adsorption capacity of [Cu(bipy)Cl₂]_n for the two dyes are much less than that of [Cu(bipy)(SO₄)]_n. Second, the adsorption capacity of [Cu(bipy)(SO₄)]_n for AO7 is much more than the exchanged sulfate ions. By theoretical calculation, if all the Cl⁻ are exchanged by AO7, the adsorption capacity should reach 2410 mg g⁻¹ (6.88 mmol g⁻¹). The experimental values illuminate that only parts of Cl⁻ anions were exchanged, which may because of the bridging bond between Cl and Cu make the Cl difficult to be exchanged. As the same kind of adsorbent, [Cu(bipy)(SO₄)]_n performs better in removing AO7 and MO. Especially for AO7, the adsorption capacity reached 3308 mg g⁻¹ (9.44 mmol g⁻¹), which is much greater than the sulfate ions. This suggests other adsorption mechanisms may account for the second adsorption step in the adsorption process of AO7 besides ion exchange.

In order to further confirm the mechanism of the second-stepwise adsorption, a cationic dye, malachite green (MG), was selected as a molecular probe to add to the system containing AO7 and [Cu(bipy)(SO₄)]_n. Fig. 4 presents the adsorption amount of MG and AO7 on [Cu(bipy)(SO₄)]_n in the condition of fixed MG (100 mg L⁻¹) mixed with varying AO7 (0–1500 mg L⁻¹). The aim of this experiment is to study the effect of the concentration of AO7 on the adsorption of MG, and the impact of the adsorption of MG on the adsorption of AO7. It shows that without AO7, MG could hardly be removed. However, with the addition of AO7, the MG could be removed efficiently. At the stage of the adsorption isotherm of MG in Fig. 4 when the C₀ of AO7 is 200 mg L⁻¹, 93.9% of MG was removed. The inset of the picture also shows their difference with or without AO7 in adsorption system. The results reveal that the [Cu(bipy)(AO7)₂]_n form by [Cu(bipy)]_n²ⁿ⁺ and AO7 could adsorb MG effectively. Fig. S2† provided the detailed information of the variation of zeta potential along with the adsorption amount. Because the zeta potential of [Cu(bipy)(AO7)₂]_n·H₂O without MG was −2.4 mV (corresponding to the picture in Fig. 4), it is rational to ignore the electrostatic effect between positive MG and [Cu(bipy)(AO7)₂]_n. The results suggest that [Cu(bipy)(AO7)₂]_n may possess the potential of hydrophobic adsorption because [Cu(bipy)(AO7)₂]_n can adsorb anionic AO7⁻ and also cationic MG with hydrophobic polycyclic moiety. Besides, the adsorption of MG did not affect the adsorption capacity of AO7 below 800 mg L⁻¹, but decreased the adsorption capacity of AO7 when the initial concentration of AO7 is 1200 mg L⁻¹ and 1500 mg L⁻¹ which belong to the second adsorption step in Fig. 2. The results indicate the adsorption of malachite green and the second step adsorption of AO7 could be possibly caused by the same reason. By the combination analysis of above experimental results, it can be drawn that the first step adsorption of AO7⁻ on [Cu(bipy)(SO₄)]_n is caused by ionic exchange and the second step is by hydrophobic interaction. In order to elucidate clearly, the speculated adsorption mechanism was expressed by Fig. 5.

Fig. 4 Adsorption behavior of malachite green on [Cu(bipy)(SO₄)]_n in the dual-dyes system with mixtures of AO7 and MG in it. The initial concentration of MG is 100 mg L⁻¹; the initial concentration of AO7 changed in the range of 0–1500 mg L⁻¹. The abscissa is set as the initial concentration of AO7. Inset is the picture of samples with initial concentration of 100 mg L⁻¹ of MG and 200 mg L⁻¹ of AO7 for comparison before (left) and after (right) adsorption.

Fig. 5 (a) The crystal cell of [Cu(bipy)(SO₄)]_n from the z direction and the first molecular layer of the crystal cell; (b) comparison of the [Cu(bipy)(SO₄)]_n adsorbed different amount of AO7; (c) the speculated adsorption process of the two-stepwise adsorption of AO7 by [Cu(bipy)(SO₄)]_n. The SO₄²⁻ on [Cu(bipy)SO₄]_n and the Na⁺ on AO7 molecule are omitted for clarity.

As shown in Fig. 5, the [Cu(bipy)(SO₄)]_n is constructed by layers of one-dimensional molecular chain of [Cu(bipy)(SO₄)]_n. At the first adsorption step of AO7, the mechanism of ion exchange takes control in the process. As more AO7 be adsorbed on the [Cu(bipy)]_n²ⁿ⁺, the one-dimensional molecular chain depart from the crystal structure, leading to the volumetric swelling of the sediment, as shown in the Fig. 5b. The adsorbed AO7 may form hydrophobic area which facilitates the second adsorption step of AO7. Considering that the two-stepwise adsorption did not happen in the adsorption process of MO, it is speculated that the π–π interaction between the naphthalene rings on the hydrophobic part of AO7 molecule may favor the second-step adsorption.

3.4. Adsorption kinetics

Adsorption isotherm curve can tell the maximal adsorption capacity of adsorbent for a specific adsorbate. Kinetic curve can show the time needed for the adsorption process reaching equilibrium. Fig. 6 shows the adsorption kinetic curves of AO7 and MO by the two adsorbents. Two classical kinetic models were used to fit the experimental data. The pseudo first-order model is referred to the Lagergren rate equation as following linear eqn (5) and non-linear eqn (6):²¹

ln(Q_e − Q_t) = ln Q_e − k₁t (5)

Q_t = Q_e[1 − exp(−k₁t)] (6)

where Q_e is the amount of dyes absorbed per gram of the adsorbent at equilibrium (mg g⁻¹), Q_t is the amount of dyes adsorbed at time t (mg g⁻¹), and k₁ is the equilibrium rate constant of pseudo first-order (h⁻¹).

Fig. 6 Adsorption kinetic curves of AO7 (a) and MO (b) by [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n.

The kinetics of adsorption was also fitted with pseudo second-order model as following linear eqn (7) and non-linear eqn (8):²²

where k₂ is the pseudo second-order rate constant ((g mg⁻¹) h⁻¹).

The results are also shown in Fig. 6; and the parameters of the two fitted models are shown in Table 3. It can be seen that, for MO, the calculated Q_e values of the pseudo-second order kinetic model fits better with the experimental data than the pseudo-first order kinetic model does. The correlation coefficients for the pseudo-second model are also better for the kinetic curves of MO. These results indicated that the adsorption processes of MO belong to the pseudo-second order kinetic model. However, the pseudo-first model is better for the fitting of the kinetic curves of AO7. Although AO7 and MO, with similar structure, are both monovalent dye, the equilibrium time for adsorption of MO is much shorter than that of AO7, which can be derived from the kinetic constant. Take the adsorption on [Cu(bipy)(SO₄)]_n for example, the pseudo second-order kinetic constant k₂ of MO is 0.0508 (g mg⁻¹) h⁻¹, being 96.6 times of that of AO7 (5.26 × 10⁻⁴ (g mg⁻¹) h⁻¹). Compare to MO, the adsorption of AO7 was constrained by some factors. By comparing their molecular structure, it is supposed that the hydrophilic hydroxyl group on AO7 is disadvantageous to its removal.

Table 3 Kinetic parameters for the adsorption of AO7 and MO by [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n

Parameters		Pseudo first-order				Pseudo second-order
		Qe (exp) (mg g^−1)	Qe (cal) (mg g^−1)	k1 (h^−1)	R^2	Qe (cal) (mg g^−1)	k2 ((g mg^−1) h^−1)	R^2
AO7	[Cu(bipy)Cl2]n	487.2	483.2	0.183	0.994	602.5	3.09 × 10^−4	0.996
	[Cu(bipy)(SO4)]n	702.3	722.6	0.179	0.996	777.7	5.26 × 10^−4	0.973
MO	[Cu(bipy)Cl2]n	703.9	675.5	3.73	0.988	757.2	0.0066	0.997
	[Cu(bipy)(SO4)]n	751.6	732.7	13.31	0.993	754.8	0.0508	0.998

---

When comparing the two adsorbents, the [Cu(bipy)(SO₄)]_n performs better in removing the dyes. The adsorption amount in equilibrium (Q_e) and kinetic constant of MO using [Cu(bipy)(SO₄)]_n as adsorbent are higher than those using [Cu(bipy)Cl₂]_n. Although the k₁ of the two adsorbents when removing AO7 are almost the same, the Q_e of [Cu(bipy)(SO₄)]_n is obviously higher than that of [Cu(bipy)Cl₂]_n. This result is in constant with the adsorption equilibrium study.

4. Conclusion

One-dimensional CP, [Cu(bipy)]_n²ⁿ⁺, through ion exchange, can adsorb AO7 and MO effectively. The anions that balance the positive charge of [Cu(bipy)]_n²ⁿ⁺ affect the character of the one-dimensional adsorbent prominently. Compare to [Cu(bipy)Cl₂]_n, the [Cu(bipy)(SO₄)]_n performed better in removing AO7 and MO. The molecular structures of the dyes also greatly affect the adsorption process. MO can be removed faster and better than AO7, as indicated by the kinetic curves and isotherms of [Cu(bipy)Cl₂]_n and [Cu(bipy)(SO₄)]_n. However, a two-stepwise adsorption isotherm was observed in the adsorption of AO7 using [Cu(bipy)(SO₄)]_n, which make the adsorption capacity of [Cu(bipy)(SO₄)]_n for AO7 reached 3308 mg g⁻¹. It is speculated that the first adsorption step was induced by ion exchange, and the second adsorption step was by hydrophobic force and π–π interaction between the AO7 in aqueous phase and the adsorbed AO7. This study revealed that one-dimensional CP, although has little surface area, can still effectively adsorb dyes like AO7 and MO from water. The adsorption mechanism of ion exchange suggests that the one-dimensional CP can be applied to the removal of other anionic dyes.

Acknowledgements

This research was supported by Nature Science Foundations of China (21107146), Nature Foundations of Guangdong Province (92510027501000005), Project of Education Bureau of Guangdong Province (cgzhzd1001), the Fundamental Research Funds for the Central Universities (121pgy20), and Science and Technology Research Programs of Guangzhou City (201510010083), and Science and Technology Key Projects of Guangdong Province (2014B020216004).

References

1. J. D. Fox and S. J. Rowan, Macromolecules, 2009, 42, 6823–6835.
2. H. Deng, S. Grunder, K. E. Cordova, C. Valente and O. M. Yaghi, Science, 2012, 336, 1018–1023.
3. N. Masciocchi, P. Cairati, L. Carlucci, G. Mezza, G. Ciani and A. Sironi, J. Chem. Soc., Dalton Trans., 1996, 2739–2746.
4. N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim and O. M. Yaghi, Science, 2003, 300, 1127–1129.
5. Z. Hasan and S. H. Jhung, J. Hazard. Mater., 2015, 283, 329–339.
6. E. Haque, J. W. Jun and S. H. Jhung, J. Hazard. Mater., 2011, 185, 507–511.
7. E. Haque, J. E. Lee, I. T. Jang, Y. K. Hwang and S. H. Jhung, J. Hazard. Mater., 2010, 181, 535–542.
8. M. Anbia, V. Hoseini and S. Sheykhi, J. Ind. Eng. Chem., 2012, 18, 1149–1152.
9. Y. Xu and R. Li, RSC Adv., 2015, 5, 44828–44834.
10. Y. Liu, G. Cui, C. Luo, L. Zhang, Y. Guo and S. Yan, RSC Adv., 2014, 4, 55162–55172.
11. T. Yao, S. Guo, C. Zeng, C. Wang and L. Zhang, J. Hazard. Mater., 2015, 292, 90–97.
12. J. Zhou, C. Tang, B. Cheng, J. Yu and M. Jaroniec, ACS Appl. Mater. Interfaces, 2012, 4, 2174–2179.
13. Y. Zheng, Y. Liu and A. Wang, Ind. Eng. Chem. Res., 2012, 51, 10079–10087.
14. R. Cheng, S. Ou, B. Xiang, Y. Li and Q. Liao, Langmuir, 2010, 26(2), 752–758.
15. C. He and X. Hu, Ind. Eng. Chem. Res., 2011, 50, 14070–14083.
16. M. Greluk and Z. Hubicki, Chem. Eng. J., 2011, 170, 184–193.
17. W. Zhang, H. Li, X. Kan, L. Dong and H. Yang, Bioresour. Technol., 2012, 117, 40–47.
18. J. Ma, F. Yu, L. Zhou, L. Jin, M. Yang and J. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 5749–5760.
19. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
20. H. M. F. Freundlich, J. Phys. Chem., 1906, 57, 385–470.
21. S. Lagergren, K. Sven. Vetenskapsakad. Handl., 1898, 24, 1–39.
22. Y. S. Ho and G. Mckay, Process Biochem., 1999, 34, 451–465.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra09830k

---