ORI GIN AL PA PER

Anhydride Functional Nanocoatings for Heavy MetalCadmium Capture and Release

T. J. Wood • J. P. S. Badyal

Received: 1 July 2013 / Accepted: 22 July 2013 / Published online: 4 August 2013� Springer Science+Business Media New York 2013

Abstract Pulsed plasmachemical deposited poly(maleic anhydride) nanolayers display

high efficiency for the capture of toxic cadmium ions from aqueous solution (down to the

low parts per billion range). Subsequent release of the isolated heavy metal species and

host media regeneration is accomplished by rinsing in weak acid solution.

Keywords Plasma polymer � Maleic anhydride � Capture and release �Water purification � Heavy metal ion

Introduction

Cadmium is used in many large-scale industrial processes including the manufacture of

batteries [1, 2], pigments [3], and plastic stabilizers [4]. A major issue in developing

countries is that cadmium is carcinogenic and leads to multiple toxic effects for humans

[5]. As a consequence, its presence in drinking water supplies (from either natural or

industrial sources [5, 6]) is highly undesirable and therefore safe cadmium levels for

consumption have been set by regulatory bodies in the low parts per billion range [7].

Previous methods for cadmium removal from wastewaters have included use of biopoly-

mers (such as polypeptides [8], starch [9], chitin [10], and chitosan [11, 12]) bacteria [13],

algae [14], fungi [15], food waste [16], metal oxides [17], ion-exchange materials [18],

activated carbon [19], or electrolysis [20]. However, these methods suffer from a lack of

specificity [21], non-recyclability [22], high cost [23], are power-intensive [20] and often

impractical for industrial level scale-up [24, 25].

In this article we describe the use of pulsed plasmachemical deposition of maleic

anhydride precursor in order to produce nanocoatings containing a high density of anhy-

dride groups which readily capture cadmium ions from water, Scheme 1. This single-step,

T. J. Wood � J. P. S. Badyal (&)Department of Chemistry, Science Laboratories, Durham University, Durham DH1 3LE, UKe-mail: j.p.badyal@durham.ac.uk

123

Plasma Chem Plasma Process (2013) 33:873–879DOI 10.1007/s11090-013-9473-5

solventless surface functionalization technique is straightforward and could be readily

translated to heavy metal ion capture and release from industrial wastewaters. Pulsed

plasmachemical deposition entails modulating an electrical discharge on the microsecond-

millisecond timescale such that precursor vapour is activated at the substrate surface (via

VUV irradiation, ion, or electron bombardment) during each microsecond on-period,

followed by conventional polymerization of the precursor carbon–carbon double bond

during each subsequent millisecond off-period [26]. This leads to polymeric coatings with

high levels of structural retention. Previous examples of functionalities that have been

deposited in this way have included: carboxylic acid [27], amine [28], cyano [29], epoxide

[30], hydroxyl [31], halide [32], thiol [33], furfuryl [34], perfluoroalkyl [35], perfluo-

romethylene [36], and trifluoromethyl [37] groups.

Experimental

Plasmachemical Deposition of Anhydride Layers

Plasmachemical deposition was carried out in an electrodeless cylindrical glass reactor

(volume of 480 cm3, base pressure of 3 9 10-3 mbar, and with a leak rate better than

2 9 10-9 mol s-1) surrounded by a copper coil (4 mm diameter, 10 turns), and enclosed

in a Faraday cage. The chamber was pumped down using a 30 L min-1 rotary pump

attached to a liquid nitrogen cold trap, and a Pirani gauge monitored system pressure.

Maleic anhydride briquettes (?99 %, Aldrich Ltd., ground into a fine powder) were loaded

Substrate

Substrate

Substrate

Cd2+

2+

(aq)

H+

Substrate

Cd2+

+Cd2+

+

2+

Scheme 1 Pulsedplasmachemical deposition ofpoly(maleic anhydride) layersfollowed by cadmium ion captureand subsequent release

874 Plasma Chem Plasma Process (2013) 33:873–879

123

into a sealable glass tube connected to the system followed by degassing through several

freeze–pump–thaw cycles. The output impedance of a 13.56 MHz radio frequency (rf)

power supply was matched to the partially ionized gas load via an L–C matching unit

connected to the copper coil. Prior to each deposition, the reactor was scrubbed using

detergent, rinsed in propan-2-ol, and dried in an oven. A silicon (100) wafer piece (Silicon

Valley Microelectronics Inc.) was then inserted into the chamber and a continuous wave air

plasma was run at 0.2 mbar pressure and 40 W power for 30 min in order to remove any

remaining trace contaminants from the chamber walls. Next, maleic anhydride precursor

vapour was allowed to purge the reactor for 5 min at a pressure of 0.2 mbar prior to

electrical discharge ignition. An optimal duty cycle of 20 ls on-period and 1,200 ls off-

period in conjunction with a peak power of 5 W was employed for pulsed plasma depo-

sition [26]. Upon plasma extinction, the precursor vapour was allowed to continue to pass

through the system for a further 3 min, and finally the chamber was evacuated back down

to base pressure prior to venting to atmospheric pressure.

Cadmium Capture and Release

Cadmium ion capture experiments entailed placing a piece of coated silicon wafer (1 cm2

area) into 2 cm3 volume of a 850 parts per billion cadmium(II) chloride (Koche-Light

Laboratories Ltd.) solution prepared using ultra high purity water (resistivity greater than

18 MX cm, organic content less than 1 ppb, Sartorius Arium 611), followed by gentle

stirring. The sample substrate was then washed in ultra high purity water for 1 h. Cadmium

ion release experiments entailed soaking the sample substrate in aqueous acetic acid

(Fisher Scientific Ltd.) solution (pH = 3.7) for 1 h in order to effect ion exchange between

immobilized Cd2? and H?(aq). Alternative acids could also be used for regeneration (e.g.

HCl).

Characterisation

Film thicknesses were measured using a spectrophotometer (nkd-6000, Aquila Instruments

Ltd.). This entailed acquisition of transmittance-reflectance curves (350–1,000 nm wave-

length range) for each sample and fitting to a Cauchy material model using a modified

Levenberg–Marquardt algorithm [38]. Film deposition rates were calculated to be

3 ± 1 nm min-1. Typical film thicknesses used for cadmium ion capture and release

studies were 100 nm.

Infrared spectra were acquired using a FTIR spectrometer (Perkin-Elmer Spectrum

One) fitted with a liquid nitrogen cooled MCT detector operating at 4 cm-1 resolution

across the 700–4,000 cm-1 range. The instrument included a variable angle reflection–

absorption accessory (Specac Ltd.) set to a grazing angle of 66� for silicon wafer substrates

and adjusted for p-polarization.

Surface elemental compositions were determined by X-ray photoelectron spectroscopy

(XPS) using a VG ESCALAB II electron spectrometer equipped with a non-monochro-

mated Mg Ka X-ray source (1,253.6 eV) and a concentric hemispherical analyser.

Photoemitted electrons were collected at a take-off angle of 20� from the substrate normal,

with electron detection in the constant analyser energy mode (CAE, pass energy = 20 eV).

Experimentally determined instrument sensitivity (multiplication) factors were taken as

C(1s) : O(1s) : Cd(3d) equals 1.00: 0.36: 0.05. All binding energies were referenced to the

C(1s) hydrocarbon peak at 285.0 eV. A linear background was subtracted from core level

Plasma Chem Plasma Process (2013) 33:873–879 875

123

spectra and then fitted using Gaussian peak shapes with a constant full-width-half-maxi-

mum (fwhm) [39, 40].

The concentration of cadmium ions present in solution was measured by atomic

absorption spectroscopy at a wavelength of 228.8 nm (Varian Spectra AA 220FS Atomic

Absorption Spectrophotometer). Cadmium equilibration standards were prepared from a

certified 1,000 mg L-1 stock solution (PlasmaCAL SCP Science).

Results

Fourier transform infrared spectra of the pulsed plasma deposited poly(maleic anhydride)

nanocoatings show distinctive carboxylic anhydride symmetric and antisymmetric C=O

stretches at 1,850 and 1,802 cm-1 respectively (denoted A and B), Fig. 1. The peak

positions are consistent with a cyclic unconjugated system, which is indicative of the

carbon–carbon double bond contained in the maleic anhydride molecule undergoing

reaction during deposition (i.e. conventional polymerization taking place) [41, 42]. These

anhydride absorbances disappear upon exposure to 870 ppb cadmium(II) chloride solution

to be replaced by carboxylic acid antisymmetric C=O stretches (1,730 cm-1, denoted C),

carboxylate antisymmetric C=O stretches (1,591 cm-1, denoted D), and carboxylate

symmetric C=O stretches (1,420 cm-1, denoted E). In addition there are carboxylic acid

dimer and monomer C–O stretches at 1,301 and 1,184 cm-1 respectively. These acid peaks

4000 3500 3000 2500 2000 1500 1000

ED

C

B

A

(g)

(f)

(e)

(d)

(c)

(b)

(a)

Tra

nsm

ittan

ce /

%

Wavenumber / cm-1

Fig. 1 Infrared spectra of pulsedplasma deposited poly(maleicanhydride) layers followingexposure to 870 ppb cadmium(II)chloride solution for: a 0 h, b 1 h,c 2 h, d 4 h, e 8 h, f 16 h, andg saturation reference using1 mol dm-3 cadmium(II)chloride solution for 1 himmersion

876 Plasma Chem Plasma Process (2013) 33:873–879

123

correlate to the anhydride groups undergoing hydrolysis in the presence of water, whilst the

carboxylate peaks correspond to ion exchange of Cd2? for H?. The carboxylate bands

become stronger with increasing exposure to the cadmium ion solution. As an absolute

reference point, the maximum signal intensity for the carboxylate features was measured

using a 1 mol dm-3 cadmium(II) chloride solution for 1 h exposure (which is shown for

comparison). Subsequent rinsing of the coatings in aqueous acetic acid gave rise to the

disappearance of the carboxylate peaks at 1,591 and 1,420 cm-1 due to the release of

cadmium ions, Fig. 1.

Atomic absorption spectroscopy analysis of the cadmium(II) chloride solutions fol-

lowing immersion of the pulsed plasma deposited poly(maleic anhydride) substrates

showed an exponential decrease of cadmium concentration versus duration of exposure,

Fig. 2. It is estimated that 50 % of the cadmium ions present in the solution can be

absorbed within 40 min and this drops to below 80 ppb after 16 h of immersion.

X-ray photoelectron spectroscopy of the pulsed plasma deposited poly(maleic anhy-

dride) nanolayers following removal from the cadmium(II) chloride solution and rinsing in

high purity water showed a concurrent increase in the surface cadmium content with length

of immersion time, Fig. 2. This cadmium concentration reached a value of around

2 atom% after 8–16 h. For the case of soaking in a reference control solution (1 mol dm-3

cadmium(II) chloride solution for 1 h), the cadmium level was measured to be 4.7 atom%.

This corresponds to a cadmium absorption capacity of 310 mg per gram of coating.

Subsequent rinsing in acetic acid solution resulted in the complete disappearance of the

cadmium XPS signal which is indicative of metal ion release/regeneration. These nano-

coatings could be reused multiple times without observing any deterioration of cadmium

ion capture and release performance.

Discussion

Previous materials and coatings employed to remove cadmium ions from solution are

reported to suffer from a lack of specificity [21] or recyclability [22]. In contrast, the

current pulsed plasma deposited poly(maleic anhydride) nanocoatings are specific to heavy

metal ions and can be regenerated by washing in mild acid solutions (similar results were

observed for zinc ions, whilst no absorption was recorded for non-toxic alkali or alkaline

earth metals and iron). Cadmium(II) complexes adopt a coordination number of 4 around

0 2 4 6 8 10 12 14 160.0

0.2

0.4

0.6

0.8

1.0

(b)

(a)

Immersion Period / h

Rem

aini

ng C

d2+

Con

cent

ratio

n / p

pm

0.0

0.5

1.0

1.5

2.0

Captured X

PS

Cd

2+ Content / atom

%Fig. 2 a Concentration ofCd2?(aq) measured by atomicabsorption spectroscopy ofcadmium(II) chloride solutionsfollowing period of immersion ofpulsed plasma depositedpoly(maleic anhydride) coatings;and b Cd2? content of thecoatings following immersion incadmium(II) chloride solutions asmeasured by X-ray photoelectronspectroscopy (excludinghydrogen)

Plasma Chem Plasma Process (2013) 33:873–879 877

123

20 % occurrence and a coordination number of 6 with about 56 % occurrence [43]. The

reason for this variety of coordination number for Cd(II) ions is due to the filled d10 orbital

[44]. Given that the plasma polymerized poly(maleic anhydride) coatings are known to

swell in water (thus providing access for large metal ions [42]), and also there exists some

degree of crosslinking [45] (which restricts metal ion coordination geometries), prefer-

ential cadmium ion capture arises due to its greater flexibility to complex (through a

combination of both 4 and 6 coordination geometries). This selectivity is verified by the

lack of capture for other metal ions (for instance alkali or alkaline earth metals, and iron).

The measured time taken to lower the cadmium ion concentration in solution by half is

around 40 min, which compares favourably with previous approaches, where the time can

take around double [25]. The maximum weight of cadmium capture per gram of coating is

310 mg, which is in the same range as biopolymers such as chitosan (100–500 mg g-1)

[46], and far better than other biological [47] and inorganic systems [48]. The high capacity

for these coatings to absorb cadmium ions in the form of metal carboxylate groups can be

attributed to the high density of carboxylic acid groups contained in the hydrated pulsed

plasma deposited layers.

The outlined pulsed plasmachemical deposition approach offers many advantages for

the fabrication of high-density carboxylic acid capture and release coatings, including

single-step fabrication, conformal, solventless, and low energy consumption [42]. In order

to make this scalable, an increase in effective surface area of the pulsed plasma deposited

nanocoatings could easily be envisaged (due to the conformality of the vapour-phase

plasma process), for instance by coating porous or high surface-area substrates (e.g.

nonwoven polypropylene cloth [49]) in combination with roll-to-roll processing.

Conclusions

Pulsed plasmachemical deposition has been shown to be an effective means for the

manufacture of anhydride coatings which specifically capture heavy metal ions. The

coatings may be regenerated by washing with acid to release the heavy metal and sub-

sequently reused for water purification.

Acknowledgments Dr. W. C. E. Schofield and Mr. M. D. West are thanked for sample preparation andatomic absorption spectroscopy analysis. T. J. Wood is grateful to Surface Innovations Ltd. for financialsupport.

References

1. Edison TA (1902) U.S. Patent No. 692,5072. David J (1995) J Power Sources 57:713. Jansen M, Letschert HP (2000) Nature 404:9804. Zerling L, Hanisch C, Junge FW (2006) Acta Hydrochim Hydrobiol 34:2345. Waisberg M, Joseph P, Hale B, Beyersmann D (2003) Toxicology 192:956. Berg JW, Burbank F (1972) Ann N Y Acad Sci 199:2497. Cotruvo J, Fawell JK, Giddings M, Jackson, Magara Y, Festo Ngowi AV, Ohanian E (2011) Cadmium

in drinking water. WHO Press, Geneva8. Prabhukumar G, Matsumoto M, Mulchandani A, Chen W (2004) Environ Sci Technol 38:31489. Jawed M, Tare V (1991) J Appl Polym Sci 42:317

10. Benguella B, Benaissa H (2002) Water Res 36:246311. Muzzarelli RAA, Sipos L (1971) Talanta 18:853

878 Plasma Chem Plasma Process (2013) 33:873–879

123

12. Masri MS, Reuter FW, Friedman M (1974) J Appl Polym Sci 18:67513. Inthorn D, Nagase H, Isaji Y, Hirata K, Miyamoto K (1996) J Ferment Bioeng 82:58014. Skowronski T, Przytocka-Jusiak M (1986) Chemosphere 15:7715. Yakup Arıca M, Kacar Y, Genc O (2001) Bioresour Technol 80:12116. Kuh SE, Kim DS (2000) Environ Technol 21:88317. Skubal LR, Meshkov NK, Rajh T, Thurnauer M (2002) J Photochem Photobiol A 148:39318. Kesraoui-Ouki S, Cheeseman C, Perry R (1993) Environ Sci Technol 27:110819. Reed BE, Matsumoto MR (1993) Sep Sci Technol 28:217920. Elsherief EA (2003) Electrochim Acta 48:26621. Fan H-J, Anderson PR (2005) Sep Purif Technol 45:6122. Wang CL, Michels PC, Dawson SC, Kitisakkul S, Baross JA, Keasling JD, Clark DS (1997) Appl

Environ Microbiol 63:407523. Escobar C, Soto-Salazar C, Toral MI (2006) J Environ Manag 81:38424. Benaıssa H (2006) J Hazard Mater 132:18925. Sternberg SPK, Dorn RW (2002) Bioresour Technol 81:24926. Ryan ME, Hynes AM, Badyal JPS (1996) Chem Mater 8:3727. Hutton SJ, Crowther JM, Badyal JPS (2000) Chem Mater 12:228228. Harris LG, Schofield WCE, Doores KJ, Davis BG, Badyal JPS (2009) J Am Chem Soc 131:775529. Tarducci C, Schofield WCE, Brewer SA, Willis C, Badyal JPS (1800) Chem Mater 2001:1330. Tarducci C, Kinmond EJ, Brewer SA, Willis C, Badyal JPS (1884) Chem Mater 2000:1231. Tarducci C, Schofield WCE, Brewer SA, Willis C, Badyal JPS (2002) Chem Mater 14:254132. Teare DOH, Barwick DC, Schofield WCE, Garrod RP, Ward LJ, Badyal JPS (2005) Langmuir

21:1142533. Schofield WCE, McGettrick J, Bradley TJ, Przyborski S, Badyal JPS (2006) J Am Chem Soc 128:228034. Tarducci C, Brewer SA, Willis C, Badyal JPS (2005) Chem Commun 40635. Coulson SR, Woodward IS, Brewer SA, Willis C, Badyal JPS (2000) Chem Mater 12:203136. Limb SJ, Gleason KK, Edell DJ, Gleason EF (1997) J Vac Sci Technol A15:181437. Wang J-H, Chen J-J, Timmons RB (1996) Chem Mater 8:221238. Lovering D (1998) NKD-6000 technical manual. Aquila Instruments, Cambridge39. Friedman RM, Hudis J, Perlman ML (1972) Phys Rev Lett 29:69240. Evans JF, Gibson JH, Moulder JF, Hammond JS, Goretzki H (1984) Fresen J Anal Chem 841:31941. Dauben WG, Epstein WW (1959) J Org Chem 24:159542. Wood TJ, Schofield WCE, Badyal JPS (2012) J Mater Chem 22:783143. Sigel H, Martin RB (1994) Chem Soc Rev 23:8344. Cotton FA, Wilkinson G, Murillo CA, Bochmann M (1999) Advanced inorganic chemistry, 6th edn.

Wiley, New York, pp 598–60045. Yasuda H (1985) Plasma polymerization. Academic Press, New York46. Rorrer GL, Hsien T-Y, Way JD (1993) Ind Eng Chem Res 32:217047. Volesky B, Praseyto I (1994) Biotechnol Bioeng 43:101048. Bentouami A, Ouali MS (2006) J Colloid Interface Sci 293:27049. Schofield WCE, Bain CD, Badyal JPS (2012) Chem Mater 24:1645

Plasma Chem Plasma Process (2013) 33:873–879 879

123