Metal Determination in Cosmetics by ICP

It is well known that heavy metals are found naturally in the environment in rocks, soil and water and therefore exist in the manufacture of pigments and other raw materials in all industries including the cosmetics industry. Some of these metals have been used as cosmetic ingredients in the past. In some cases, measures have been implemented to reduce the amount of heavy metals to which users are exposed, including prohibiting their use in cosmetics. Lead,arsenic, cadmium, mercury, antimony and chromium are the main heavy metal ingredients prohibited in cosmetics in most countries. Cosmetic manufactures have the responsibility to ensure their products do not contain any of these metals.

To read the full application note, please login or create a free account.

Introduction

It is well known that heavy metals are found naturally in the environment in rocks, soil and water, and therefore exist in the manufacture of pigments and other raw materials in all industries including the cosmetics industry. Some of these metals have been used as cosmetic ingredients in the past. In some cases, measures have been implemented to reduce the amount of heavy metals to which users are exposed, including prohibiting their use in cosmetics. Lead, arsenic, cadmium, mercury, antimony and chromium are the main heavy metal ingredients prohibited in cosmetics in most countries. Cosmetic manufacturers have the responsibility to ensure their products do not contain any of these metals.

Metal determination in samples is often done by ICP or ICPMS and it requires the samples to be in dissolved form. Commonly, acid digestions are the method of choice but seeing as many samples contain an important amount of silica, alumina and magnesium, some aggressive acids, such as HF must be used to obtain full dissolution. Furthermore, the quickest microwave methods can take up to 3 hours.

Knowing the risks associated with the use of HF, many laboratories look for alternative methods for obtaining full dissolution of their samples while optimizing the uptime and productivity.

The proposed methods describe alternative ways to determine the concentration of elements of interest as well as confirm the absence of heavy metals contained in cosmetics using borate fusions and peroxide fusions for ICP-OES determination.

Method

1. Sample Preparation

1.1 - Ashing
Due to their high organic and water content, each cosmetic sample and standard was submitted to an ashing procedure.
• Porcelain crucibles
• 550 °C until constant weight
• LOI is calculated for final analysis

1.2 - Borate fusion Method
Claisse® TheOx-DS® 6 position electric Fluxer
• Pt-Au (95 % / 5%) crucibles
• 34.83 / 64.67 / 0,5 (Li₂B₄O₇ / LiBO₂ / LiBr) flux
• 10 minutes of heating at 1050 °C
• 4 minutes of dissolution in 10 % HNO₃ v / v

1.3 - Peroxide fusion Method
Claisse® TheOx-DS® 6 position electric Fluxer
• zirconium crucibles
• sodium peroxide flux
• 7 minutes of heating at 700 °C
• 4 minutes of cooling
• Dissolution in 20 % HNO₃ / HCl v / v

2. ICP-OES

2.1 – PerkinElmer® Optima 7300DV

Table 1: Optima 7300 DV operating parameters
Plasma flow rate (Ar)	16.0 L/min
Auxiliary gas flow rate	0.4 L/min
Nebulizer flow rate	0.8 L/min
Sample flow rate	1.0 mL/min
Rinse	2.5 mL/min (2 min < 10 ppm; add. 90 sec for > 10ppm)
Shear gas pressure	100 PSI
RF Power	1500 W

2.2 – Samples

Table 2: Reference Materials and Samples Used to Validate the Developed Method with their calculated Losses on ignition (LOI )
Sample	Supplier	LOI (%)
DC60132 (Talc CRM)	NCS	10
SDC-1 (Mica CRM)	USGS	2
Sample 1 (647G-08-8008)	Cosmetic Industry	63
Sample 2 (6LX5-11-8001)	Cosmetic Industry	88
Sample 3 (6MPY-01-8001)	Cosmetic Industry	24
Sample 4 (6MNY-18-8001)	Cosmetic Industry	78
Sample 5 (9995-12-2921)	Cosmetic Industry	55

2.3 – Analytical Method

Table 3: Analytes of interest with selected wavelengths, detection limits and viewing modes for the peroxide fusion method and for the borate fusion method
Element	Wavelength (nm)	Viewing mode	MDL (mg/L)	MDL (mg/L)
			Peroxide Fusion Method	Borate Fusion Method
Al	394.401	Axial	0.2	0.2
Ba	413.065	Axial	0.1	0.1
Ca	315.887	Axial	3	0.1
Cd	214.440	Axial	0.003	0.002
Co	230.786	Axial	0.002	0.01
Cr	267.716	Axial	0.002	0.03
Fe	238.204	Radial	0.8	0.8
K	766.490	Axial	3	0.5
Mg	279.077	Axial	0.2	0.3
Mn	257.610	Axial	0.3	0.3
Mo	204.597	Axial	0.05	0.05
Na	589.592	Axial	ND*	0.4
Ni	231.604	Axial	0.001	0.008
P	213.617	Axial	0.2	0.2
Pb	220.353	Axial	0.07	0.03
Si	251.611	Axial	3	3
Sr	460.733	Axial	0.04	0.005
Ti	334.940	Radial	0.5	0.5
Zn	206.200	Radial	0.3	0.3

^{*Not determined}

Results

1. Peroxide Fusion Method

Table 4.1: Accuracy and precision measurements on SDC-1 using the peroxide fusion method
Element	Wavelength (nm)	Average experimental values (%) n = 10	Certified values (%)	Accuracy (%)	Precision (%)
Al	394.401	7.9	8.3622	95	2
Ca	315.887	BQL	1.0006	(104)	(12)
Fe	238.204	4.6	4.4204	105	3
K	766.490	2.6	2.7229	96	3
Mg	279.077	0.99	1.0191	96	1
Mn	257.610	BQL	0.0880	(90)	(1)
P	213.617	BQL	0.0698	(92)	(3)
Si	251.611	34	30.7572	109	2
Ti	334.940	0.61	0.6055	100	3

Table 4.2: Accuracy and precision measurements on DC60132 using the peroxide fusion method
Element	Wavelength (nm)	Average experimental values (%) n = 10	Certified values (%)	Accuracy (%)	Precision (%)
Al	394.401	4.0	4.0329	99	3
Ca	315.887	BQL	1.7081	(104)	(6)
Fe	238.204	1.7	1.8465	94	2
K	238.204	BQL	0.0108	-	-
Mg	279.077	17	17.7896	96	2
Mn	257.610	BQL	0.0183	(82)	(1)
P	213.617	BQL	0.048	(110)	(7)
Si	251.611	25	22.3013	114	1
Ti	334.940	0.32	0.3117	103	4

Table 4.3: Recovery results on pre-fusion spikes (n=5) using the peroxide fusion method
^Element	^{Wavelength (nm)}	^{Sample 1 (%)}	^Sample ^2 (%)	^{Sample 3 (%)}	^{Sample 4 (%)}	^{Sample 5 (%)}	^{DC60132 (%)}	^SDC-1 (%)
Ba	413.065	108	102	107	106	104	110	87
Cd	214.440	108	103	95	97	102	100	98
Co	230.786	94	91	97	96	98	100	97
Cr	267.716	95	110	111	111	114	115	106
Mo	204.597	85	105	101	104	107	102	86
Ni	231.604	99	86	99	100	102	101	84
Pb	220.353	89	86	97	97	100	93	85
Sr	460.733	100	96	104	105	112	94	86
Zn	206.200	103	101	112	92	110	97	91

2. Borate Fusion Method

Table 5.1: Accuracy and precision measurements on SDC-1 using the borate fusion method
Element	Wavelength (nm)	Average experimental values (%) n = 10	Certified values (%)	Certified values (%)	Precision (%)
Al	394.401	8.3	8.322	100	2
Ca	315.887	1.1	1.0006	105	3
Fe	238.204	4.6	4.4204	103	3
K	766.490	2.6	2.7229	94	2
Mg	279.077	1.0	1.0191	101	2
Mn	257.610	BQL	0.088	(84)	(2)
Na	589.592	1.1	1.5208	75	3
P	213.617	BQL	0.0698	(75)	(5)
Si	251.611	31	0.0698	101	2
Ti	334.940	0.6	0.6055	98	3

Table 5.2: Accuracy and precision measurements on DC60132 using the borate fusion method
Element	Wavelength (nm)	Average experimental values (%) n = 10	Certified values (%)	Accuracy (%)	Precision (%)
Al	394.401	4.3	4.0329	105	2
Ca	315.887	1.9	1.7081	114	3
Fe	238.204	1.8	1.8465	96	3
K	766.490	BDL	0.0108	-	-
Mg	279.077	18	17.7896	103	2
Mn	257.610	BDL	0.0183	(76)	(2)
Na	589.592	BDL	0.0182	-	-
P	213.617	BDL	0.0480	(92)	(3)
Si	251.611	24	22.3013	108	2
Ti	334.940	0.36	0.3117	102	2

Table 5.3: Recovery results on pre-fusion spikes (n=5) using the borate fusion method
^Element	^{Wavelength (nm)}	^{Sample 1 (%)}	^{Sample 2 (%)}	^{Sample 3 (%)}	^{Sample 4 (%)}	^{Sample 5 (%)}	^{DC60132 (%)}	^SDC-1 (%)
Ba	413.065	99	92	101	102	100	95	99
Cd	214.440	87	88	98	87	114	113	87
Co	230.786	100	105	100	105	101	101	101
Cr	267.716	108	111	102	94	111	108	110
Mo	204.597	102	106	104	106	104	103	98
Ni	231.604	98	104	99	101	104	100	101
Pb	220.353	86	87	91	85	93	86	77
Sr	460.733	95	98	86	96	87	91	84
Zn	460.733	100	94	102	102	108	113	111

Discussion

The following criteria were taken into consideration in selecting the elemental wavelengths:
(a) the freedom from spectral interferences
(b) the different sensitivities and expected concentration in the samples.

The most sensitive line was not always used in order to avoid spectral interferences and to remain in the linear range. Observed interferences were compensated for by modifying the processing parameters (e.g. adjusting the background correction points, applying multicomponent spectral fittings (MSF) or inter-elemental corrections (IEC)). To compensate for instrumental signal variations, Ge (209.419) was added to every solution as an internal standard.
Since some elements were present as majors and others were present at trace levels, it was decided that the method validations would be done following two (2) avenues. For both dissolution methods, the accuracy and precision were evaluated on the high-concentration elements whereas spike recoveries were performed on the samples and the CRMs (Tables 4.3 and 5.3) to monitor the low concentration elements. The accuracy was determined by calculating the elemental recovery of certified reference materials (CRMs). The precision was determined by preparing and measuring 10 replicates of various CRMs. The results for each CRM are presented in Tables 4.1, 4.2, 5.1 and 5.2. The accuracy, precision and recovery results obtained demonstrate that the developed methods both perform very well.

Conclusion

Peroxide fusions and borate fusions combined with the simultaneous ICP-OES (Optima 7300 DV) have the analytical capabilities to perform the analysis of many high and low concentration elements in cosmetics following a controlled ashing procedure. Metal components were measured in a variety of cosmetic samples and reference materials, demonstrating good accuracy, precision and recovery. The dual dissolution capabilities of TheOx® allowed us to validate two dissolution methods and find that both peroxide fusions and borate fusions can be used for metal determination in cosmetic products. The results obtained show that borate fusions have a slight advantage due to the lower MDLs for the majority of the elements. Finally, both methods confirm the absence of heavy metals in the cosmetic samples.

Acknowledgments: Sylvain Roy, Lab technician | Claisse and Aaron Hineman, Product Specialist | PerkinElmer®

Metal determination in cosmetics by ICP