Ref 1
Ref 1
Ref 1
https://doi.org/10.1007/s00604-018-2737-2
ORIGINAL PAPER
Abstract
Isotropic silver nanoparticles (iAg NPs) can be easily prepared at low costs, have a low electrochemical potential and high
extinction coefficient. An effective colorimetric assay for H2O2 is reported here based on the finding that H2O2 can induce the
shape transformation of citrate-capped iAg NPs with the help of citrate. The substantial shape variation affords an apparent
surface plasmon resonance (SPR) shift, accompanied by a vivid color change from light yellow to mauve. The color change can
be observed visually if the concentration of H2O2 is 2 μM or higher. A good linear relationship was obtained over the concen-
tration range of 0.2–32 μM with a limit of detection of 90 nM. By making use of glucose oxidase, the method is further extended
to glucose detection. Glucose at a concentration as low as 10 μM can be well determined with bare eyes. Benefitting from the
high selectivity, the detection of glucose in human serum is realized, and the results are in good agreement with those provided by
a clinical analyzer.
Keywords Surface plasmon resonance . Absorption . Nanomaterials . Colorimetry . Shape transformation . Hydrogen peroxide .
Colorimetric and Ag nanoparticles
positive signals from autoaggregation pose a significant chal- assay for H2O2 based on citrate-promoted H2O2-mediated
lenge in complex system applications. Hence, non-aggregation shape transformation of iAg NPs. We discovered that cit-
colorimetric sensors are gaining increasing attentions. In non- rate can promote iAg NPs collapse into anisotropic shape
aggregation colorimetric sensors, a nanomaterial solution re- in the presence of trace amount of H2O2, resulting in evi-
mains in a monodisperse state, but either the SPR intensity or dent red-shift SPR and color change. Notably, this process
the SPR position is changed upon the addition of H2O2 varies from previous studies that smoothen anisotropic Au/
[27–29]. Tuning the size of the Au/Ag nanoparticles (Au/Ag Ag nanomaterials, but directly roughen iAg NPs into an-
NPs) to change their absorption intensity and altering the shape isotropic shape. The substantial variation in the shape of
of Au/Ag nanomaterials to induce SPR shifts are two of the iAg NPs in the presence of H2O2 affords an apparent SPR
most effective design rationales [30–32]. Compared with the shift, offering this method a high sensitivity, whereas the
intensity mode, the SPR shift assays are preferred because these moderate reactivity of iAg NPs ensured the selectivity of
methods exhibit more vivid color variations, thereby offering the detection. We suppose that this study will offer a new
increased sensitivity and reliability. Currently available SPR method for designing colorimetric assays and promoting
shift-based assays are primarily based on anisotropic Au/Ag the application of i-Ag NPs in trace analysis.
nanomaterials (ai-Au/Ag NMs) which possess some highly re-
active spots or faces. The aspect ratio of these anisotropic nano-
particles decreased upon reacting with H2O2, leading to a blue- Materials and methods
shifted LSPR. For instance, it has been reported that gold nano-
rods (Au NRs) can be selectively etched by H2O2 in the longi- Materials
tudinal direction, giving rise to a visual color change; however,
because of the high electrochemical potential of Au, either high All reagents were of analytical grade and used without further
concentrations of H2O2 or catalytic species are required in these purification. Trisodium citrate was bought from Alfa Aesar.
assays [33, 34]. Ag triangle nanoprisms (Ag TNPs) are another Citrate functionalized isotropic silver nanoparticle (cit-iAg
kind of anisotropic nanomaterials that are commonly used in NPs, 50 nm), polyvinyl pyrrolidone-protected iAg NPs
H2O2 detection. Xia demonstrated that sharp tips and side faces (pvp-iAg NPs, 50 nm), and glucose oxidase (GOx,
of Ag TNPs can be etched by H2O2, inducing a triangle to 163,314 U/g) were purchased from Sigma-Aldrich
round shape transformation, affording a significant LSPR blue (Germany, https://www.sigmaaldrich.com/china-mainland.
shift, accompanied by a blue to mauve color changes [35–37]. html). Hydrogen peroxide (H2O2, 30 wt%), amino acids,
The highly reactive edges/tips of the Ag TNPs offer these glucose, fructose, lactose, sucrose, maltose, mannose were
methods outstanding sensitivity. However, selectivity becomes purchased from Aladdin Chemical Company (China, http://
a major challenge because anions, such as Cl−, Br−, I−, H2PO4−, www.aladdin-e.com/). The metal salt inducing NaF, NaCl,
and SCN−, can also react with these highly reactive spots and NaH 2 PO 4 , KBr, KSCN and KI were purchased from
faces [38–40]. Sinopharm Chemical Reagent Co (China, http://www.
Different from ai-Au/Ag NMs, isotropic Ag NPs (iAg instrument.com.cn/netshow/SH101458/). Deionized water
NPs) are not only easier in preparation, and exhibit higher was used in all experiments.
stability, but also, most importantly, exhibit moderate reac-
tivity that offers an improved selectivity, making them ide- Instruments
al candidates for colorimetric H 2O 2 detection. Despite
these favorable properties, non-aggregation colorimetric UV − vis spectra were recorded on a Lambda 365 UV spec-
sensor for H2O2 based on the SPR shift of mere iAg NPs trometer (PerkinElmer, America). Scanning electron micro-
has rarely been reported. One of the main reasons is that it scope (SEM) images were taken with a S-8010 scanning elec-
is more difficulty in inducing apparent SPR shift of iAg tron microscope (Hitachi, Japan) at an accelerating voltage of
NPs than that of ai-Au/Ag NMs because of the scarcity of 10kv. The hydrodynamic diameters and size distributions of
specialized reactive spots in iAg NPs. In fact, current iAg the Ag NPs suspensions at 25 °C were determined by dynamic
NPs-based H2O2 assay are generally operates on the size light scattering (DLS) (Brookhaven, New York). Origin 8.0
change-induced absorption intensity variation [27, 30, 31]. and excel 2010 is used to process the data. As for UV, DLS
Although previous reports have successfully realized the and SEM analysis, the corresponding built-in software of the
anisotropic growth of Ag seeds, the requirement of high instrument is used.
concentration of H2O2, H2O2 injection rate, mixing effi-
ciency, and the initial Ag concentration-dependent shape Detection of H2O2
transformation efficiency prevent the applicability of these
strategies for H2O2 detection purpose [41]. We reported A typical H2O2 detection procedure by using the cit-iAg NPs
herein for the first time a non-aggregation colorimetric was conducted as follows: 0.3 mL of the cit-iAg NPs solution
Microchim Acta9 1:58 ) 102( Page 3 of 7 19
was diluted with 2.7 mL water. Subsequently, citrate was The SPR response toward F−, Cl−, H2PO4−, Br−, SCN− and I−
added to a final concentration of 6 mM. Afterward, 3 μL of was tested by adding 20 μM these anions respectively in the
H2O2 solution with different concentrations was added to the absence of H2O2. To detect glucose in human serum, human
above mixture. The resulting mixture was incubated at room serum samples were obtained from a local hospital and centri-
temperature for 60 min. Finally, the absorbance spectra of the fuged using an Amicon Ultrafilter with a 1000 molecular
solution were obtained by a UV–vis spectrometer. weight cutoff at 5000 rpm for 10 min. The samples were
spiked with different concentrations of glucose. The resulting
mixture (200 μL) were then incubated with 1.8 mL GOx so-
Procedures for determination of glucose lution (1.5 mg/mL, buffered with 50 mM sodium acetate,
pH 5.1) and incubated at 37 °C for 30 min. Then 10 μL of
The procedure for glucose detection was as follows: 200 μL of the above mixture were added to the colloidal cit-iAg NPs
glucose solutions with different concentrations were added into (3 mL) and incubated at 37 °C for 60 min.
1.8 mL of GOx solution (1.5 mg/mL, buffered with 50 mM
sodium acetate, pH 5.1) and incubated at 37 °C for 30 min.
Then 10 μL of the above mixture was added to the cit-iAg NPs
(3 ml) and incubated at room temperature (25 °C) for 60 min. Results and discussion
Finally, the absorbance of the solutions was measured at room
temperature. All absorbance detections were performed under H2O2 sensing mechanism
the same conditions. The selectivity toward glucose was eval-
uated by using potentially interfering substances that are com- According to previous reports, H2O2 can oxidize Ag NPs and
monly present in the serum. These substances include 16 vari- decrease their size and corresponding absorbance intensity
eties of amino acids (cysteine, phenylalanine, glutamate, tryp- through the following reactions: Ag NP + 2H2O2 → Ag++
tophan, valine, isoleucine, glycine, lysine, threonine, histidine, O2·− + 2H2O; [27, 30]. Nevertheless, we discovered that when
asparagine, proline, alanine, serine, leucine, and arginine), 20 μΜ H2O2 was added into the citrate-capped isotropic Ag
ascorbic acid, and dopamine. Moreover, the selectivity toward NPs (cit-iAg NPs) solution, not only the absorption intensity
glucose analogs was evaluated by using sucrose, mannose, of the cit-iAg NPs decreased, but the maximum absorption
cellobiose, lactose, fructose, and maltose. Assay procedures peak shifted to the long wavelength (Fig. S1A). Importantly,
for the selectivity study were all the same as those described we found that this red shift can be further enhanced by exter-
in the glucose assay section except the presence of glucose. nal free citrate molecules. As demonstrated in Fig. S1B, the
citrate was added (Fig. S1D). This result confirmed that both
Glucose the free and capping citrates play crucial roles in the H2O2
Glucose acid
response process. We deduced that the SPR shift of the cit-
iAg NPs in response to H2O2 is a manifestation of morpho-
logical transformation occurred to cit-iAg NPs. To reveal in-
sight into this process, the morphologies of the cit-iAg NPs in
O2 H2O2
GOx the absence and presence of different amount of H2O2 were
characterized by scanning electron microscopy (SEM). Fig.1
clearly shows that cit-iAg NPs were gradually sculptured by
H2O2, leading to anisotropic, flawed structure. Furthermore,
DLS analysis reveal that the cit-iAg NPs solution remain
Citrate cit-iAg NPs monodispersed even after treating with 20 μΜ H2O2 (Fig.
S2). Therefore, the possibility that the red shift in SPR was
induced by cit-iAg NPs aggregation can be discounted.
Scheme 1 Proposed sensing mechanism for the detection of H2O2 and
glucose According to previous reports, citrate can serve as a shape-
directing agent and play critical roles in the formation of an-
isotropic Au/Ag nanomaterials [42, 43]. We therefore ascribed
the effective SPR red shift of the cit-iAg NPs in the presence
Δλ increased significantly with increasing citrate concentra- of H2O2 to the citrate-promoted H2O2 anisotropic sculpturing
tions from 0 to 6 mM, reached the maximum at 6 mM of the cit-iAg NPs (Scheme 1). The reduced diameters re-
(Δλ = 60 nm), and then slightly decreased at higher citrate vealed by the DLS analysis explained the decreased SPR
concentrations. However, no obvious SPR shift was observed intensity.
when treating cit-iAg NPs with 6 mM citrate alone (Fig. S1C),
suggesting that the SPR shift was caused by H2O2. Based on Sensing performance to H2O2
the above results, the concentration of 6 mM citrate was se-
lected as an optimal condition in the following assay. Also, it To achieve the most remarkable performance for H2O2 detec-
was noted that when the cit-iAg NPs were replaced with pvp- tion, we examined the time-dependent SPR shift on exposure
iAg NPs, the shifted Δλ can be ignored, even when 6 mM to H2O2 (60 μM). Fig. S3 shows that the corresponding SPR
peak shifts considerably fast in the first 10 min and then grad- a µM: 0 10 20 40 80
ually decreases to a saturation value after 60 min. Thus, the
reaction time was set at 60 min in subsequent experiments.
Under optimized conditions, the sensing strategy was evalu-
ated with respect to its performance in detecting H2O2. A
distinct color change was observed with increased glucose
concentrations. The lowest H2O2 concentration observed by
the bare eye was low at approximately 2 μM (Fig. 2a). For the
quantitative detection of H2O2, the absorbance spectra in the
presence of various concentrations of H2O2 were monitored.
As shown in Fig. 2b, the maximum absorbance band shifted
gradually with increasing H2O2 concentrations. Plotting Δλ
as a function of H2O2 concentration showed a good linear
relationship over the concentration range of 0.2–32 μM with
a limit of detection (LOD) of 90 nM at a signal-to-noise ratio
of 3 (Fig. 2c).
b
Table 1 Comparison of different colorimetric methods for H2O2 and glucose determination
MOFs, metal organic framework; TMB, 3,3,5,5-tetramethylbenzidine; BDBA, benzene-1,4-diboronic acid; OPD, o-phenylenediamine; HRP, horseradish
peroxidase; PVA, poly(vinylalcohol); ×, not given
colloidal cit-iAg NPs (Fig. S6). This result is attributed to the complications. To validate the practical application assay in
moderated reactivity of iAg NPs as compared with that of Ag biological samples, the concentrations of glucose in human
TNPs. serum samples were determined by the method of standard
addition. Considering the normal fasting blood glucose level
in the healthy human blood (3.9–6.1 mmol/L) and the high
Determination of glucose in human serum
sensitivity of our method, 200 μL of serum should well satisfy
the analysis need in each glucose measurement. Human serum
Blood glucose level is closely associated with diabetes and
samples were obtained from a local hospital and diluted and
hypoglycemia. Thus, frequent monitoring and strict control
spiked with different concentrations of standard glucose solu-
of blood sugar levels are typically requisite for effective man-
tions after centrifugation treatment. Upon adding this mixture
agement of diabetes mellitus and reduction of associated
to the solution containing cit-iAg NPs, a prolonged response
time was observed as compared with that of glucose alone
17. Sepulveda B, Angelome PC, Lechuga LM, Liz-Marzan LM (2009) 31. Tashkhourian J, Hormozi-Nezhad MR, Khodaveisi J, Dashti R
LSPR-based nanobiosensors. Nano Today 4(3):244–251 (2011) A novel photometric glucose biosensor based on decoloriz-
18. Anker JN, Hall WP, Lyandres O, Shah NC, Zhao J, Van Duyne RP ing of silver nanoparticles. Sensor Actuats B 158(1):185–189
(2008) Biosensing with plasmonic nanosensors. Nat Mater 7(6): 32. Zhang Z, Chen Z, Cheng F, Zhang Y, Chen L (2017) Highly sen-
442–453 sitive on-site detection of glucose in human urine with naked eye
19. Zeng S, Yong KT, Roy I, Dinh XQ, Yu X, Luan F (2011) A review based on enzymatic-like reaction mediated etching of gold nano-
on functionalized gold nanoparticles for biosensing applications. rods. Biosens bioelectron 89. Part 2: 932–936
Plasmonics 6(3):491–506 33. Liu X, Zhang S, Tan P, Zhou J, Huang Y, Nie Z, Yao S (2013) A
20. Lin T, Li Z, Song Z, Chen H, Guo L, Fu F, Wu Z (2016) Visual and plasmonic blood glucose monitor based on enzymatic etching of
colorimetric detection of p-aminophenol in environmental water gold nanorods. Chem Commun 49(18):1856–1858
and human urine samples based on anisotropic growth of ag nano- 34. Saa L, Coronado-Puchau M, Pavlov V, Liz-Marzan LM (2014)
shells on au nanorods. Talanta 148:62–68 Enzymatic etching of gold nanorods by horseradish peroxidase
21. Lin T, Wu Y, Li Z, Song Z, Guo L, Fu F (2016) Visual monitoring and application to blood glucose detection. Nano 6(13):7405–
of food spoilage based on hydrolysis-induced silver metallization of 7409
au Nanorods. Anal Chem 88(22):11022–11027 35. Xia Y, Ye J, Tan K, Wang J, Yang G (2013) Colorimetric visuali-
22. Wu ZS, Zhang SB, Guo MM, Chen CR, Shen GL, Yu RQ (2007) zation of glucose at the submicromole level in serum by a homog-
Homogeneous, unmodified gold nanoparticle-based colorimetric enous silver Nanoprism–glucose oxidase system. Anal Chem
assay of hydrogen peroxide. Anal Chim Acta 584(1):122–128 85(13):6241–6247
23. Radhakumary C, Sreenivasan K (2011) Naked eye detection of 36. Chen ZB, Zhang CM, Wu Q, Li K, Tan LL (2015) Application of
glucose in urine using glucose oxidase immobilized gold nanopar- triangular silver nanoplates for colorimetric detection of H2O2.
ticles. Anal Chem 83(7):2829–2833 Senson Actuators B 220:314–317
24. Wang F, Liu XQ, Lu CH, Willner I (2013) Cysteine-mediated ag-
37. Cai T, Gao Y, Yan J, Wu Y, Di J (2017) Visual detection of glucose
gregation of au nanoparticles: the development of a H2O2 sensor
using triangular silver nanoplates and gold nanoparticles. RSC Adv
and oxidase-based biosensors. ACS Nano 7(8):7278–7286
7(46):29122–29128
25. Yang YC, Tseng WL (2016) 1,4-Benzenediboronic-acid-induced
aggregation of gold nanoparticles: application to hydrogen peroxide 38. Yang XH, Ling J, Peng J, Cao QE, Ding ZT, Bian LC (2013) A
detection and biotin-Avidin-mediated immunoassay with naked- colorimetric method for highly sensitive and accurate detection of
eye detection. Anal Chem 88(10):5355–5362 iodide by finding the critical color in a color change process using
26. Vilela D, Cristina Gonzalez M, Escarpa A (2012) Sensing colori- silver triangular nanoplates. Anal Chim Acta 798:74–81
metric approaches based on gold and silver nanoparticles aggrega- 39. An J, Tang B, Zheng X, Zhou J, Dong F, Xu S, Wang Y, Zhao B, Xu
tion: chemical creativity behind the assay. A review. Anal Chim W (2008) Sculpturing effect of chloride ions in shape transforma-
Acta 751:24–43 tion from triangular to Discal silver Nanoplates. J Phys Chem C
27. Wu C, Xia G, Sun J, Song R (2015) Synthesis and anisotropic self- 112(39):15176–15182
assembly of ag nanoparticles immobilized by the Pluronic F127 40. Jiang XC, Yu AB (2008) Silver Nanoplates: a highly sensitive ma-
triblock copolymer for colorimetric detection of H2O2. RSC Adv terial toward inorganic anions. Langmuir 24(8):4300–4309
5(118):97648–97657 41. Parnklang T, Lertvachirapaiboon C, Pienpinijtham P, Wongravee
28. Xiong YM, Zhang YY, Rong PF, Yang J, Wang W, Liu DB (2015) K, Thammacharoen C, Ekgasit S (2013) H2O2-triggered shape
A high-throughput colorimetric assay for glucose detection based transformation of silver nanospheres to nanoprisms with control-
on glucose oxidase-catalyzed enlargement of gold nanoparticles. lable longitudinal SPR wavelengths. RSC Adv 3(31):12886–
Nano 7(38):15584–15588 12894
29. Lu S, Zhang X, Chen L, Yang P (2018) Colormetric determination 42. Xue C, Métraux GS, Millstone JE, Mirkin CA (2008) Mechanistic
of ferrous ion via morphology transition of gold Nanorods. study of Photomediated triangular silver Nanoprism growth. J Am
Microchim Acta 185:76 Chem Soc 130(26):8337–8344
30. Filippo E, Serra A, Manno D (2009) Poly(vinyl alcohol) capped 43. Millstone JE, Métraux GS, Mirkin CA (2006) Controlling the edge
silver nanoparticels as localized surface plasmon resonance-based length of gold Nanoprisms via a seed-mediated approach. Adv
hydrogen peroxide sensor. Sensors Actuators B 138:625–630 Funct Mater 16(9):1209–1214