391
Clinica Chimica Acta, 70 (1976) 391-398 @ Elsevier Scientific Publishing Company,
Amsterdam
- Printed
in The Netherlands
CCA 7835
DETERMINATION OF ZINC IN FINGERNAILS ABSORPTION SPECTROSCOPY
ARTHUR
SOHLER,
PATRICK
WOLCOTT
Brain Bio Center, 1225 State Road,,Princeton, (Received
February
BY NON-FLAME
ATOMIC
and CARL C. PFEIFFER N.J. 08540
(U.S.A.)
11,1976)
Summary The determination of Zn in fingernails directly using the graphite furnace presented certain difficulties due to the anomalous behavior of the analyte in the furnace. The appearance of two peaks which were due to Zn and not to any background interference was noted. The Zn value obtained by adding the area of these two peaks compared fairly well with Zn levels determined by wet ashing and subsequent determination either in the furnace or flame. Wet ashed samples gave only a single peak. It was possible to produce a model of the phenomenon with various Zn salts in a non aqueous matrix. Under these conditions ZnS04 and ZnO gave a discretely different peak from ZnClz or metallic Zn. Several tissues such as serum, whole blood, cuticle, and hair were examined for multiple peak formation. Direct determination of Zn in fingernails with the graphite furnace is possible for certain applications such as the determination of Zn levels of white spots in fingernails. For this purpose it is possible to use a sample size as small as 20 pg using the 2138 Zn line. This allows one to run several determinations on a single white spot. However, where sample size is not a limitation, wet ash digestion prior to determination in the furnace is probably the preferred procedure.
Introduction The possibility of using trace metal analysis of readily available tissues such as nail or hair for diagnostic purposes has been considered by a number of investigators. Harrison and Tyree [l] have proposed the determination of Cu in nails for diagnosis of cystic fibrosis. Robson and Brooks [ 23 have demonstrated an increase in the sodium and calcium concentration of nails from children suffering from Kwashiorkor. The occurrence of white spots in the fingernails of many children, teenagers,
392
and a few adults and their possible ~lationship to a Zn deficiency was reported by Pfeiffer and Jenney [3]. Previously Muerhcke [4] had ascribed the phenomenon to a low serum albumin level. His largest group of patients suffered from nephrosis while the next most prevalent group suffered from hepatic cirrhosis. All had low serum albumin levels. A major portion of serum zinc, about 70% is bound to albumin so that zinc depletion could be responsible for the white spots. In addition, Pfeiffer reports that the white spots disappear in patients who are on zinc therapy. The present report is concerned with an attempt to determine whether there is a zinc deficiency in the white spots of nails per se or is the role of zinc in white spot formation more indirect. The use of flameless atomic absorption was necessitated by limitations of sample size and matrix and the sensitivities of available zinc lines. Two methods of sample introduction into the graphite furnace were compared. Because of the nature of the matrix, several problems were encountered which may have practical and theoretical implications with regard to flameless atomic absorption. Materials and methods Reagents and glassware Reagents were prepared and the final rinses of glassware were made with twice glass distilled water which was subsequently passed through a Barnstead research model mixed bed ion exchange resin. Baker analytical reagent grade nitric acid was used for the digestions This gave a lower blank for Zn than their Ultrex grade nitric acid. Standards were prepared by appropriate dilution of Zn standard 1000 ppm, Fisher certified for atomic absorption. Glassware for trace metal analysis was specially cleaned by soaking in 50% nitric acid for 30 min, subsequently rinsed with the double distilled, deionized water and dried in an oven. Collection and sample ~~~dl~~g Nail clippings were obtained free of gross cont~ination such as nail polish. Clippings were washed twice with acetone, and dried at 100°C for 10 min and allowed to cool in desiccator. White spots and adjacent normal nail areas were cut up into appropriate pieces for analysis, and weighed on a Sierra electromagnetic ultra microbalance. Acid digests of approximately 1 mg nail samples were prepared by adding 0.04 ml concentrated nitric acid in a 12 mm X 75 mm Pyrex test tube. The tube was covered with a glass marble and heated at 65°C for 1 h. Tubes were cooled and diluted with 0.96 ml of distilled deionized water. lo+1 samples of digests were introduced into the furnace using an Eppendorf micro pipetter. Solid nail samples (IO-20 pg) were weighed in tantalum boats on the ultra microbalance and introduced into the furnace with the Perkin-Elmer sampling spoon.
Apparatus A Perkin-Elmer
503 Atomic Absorption
Spectrophotometer
with background
393
corrector in conjunction with an HGA 2000 graphite furnace was employed in this study. A Perkin-Elmer Model 165 chart recorder was used to record the absorbance signal. The peak areas were determined by cutting out the peak and weighing it. Chart speed was usually 60 mm/min although occasionally 240 mm/min was used. The instrumental parameters were as follows: Zn was determined with the 2138 line using a Zn, Cu, Mn, Fe multielement lamp. Furnace parameters were: drying, 125”C, 30 s (seconds) ashing, 5OO”C, 30 s and atomization 1900°C for 10 s. Nitrogen flow rate was 5. On the spectrophotometer the slit setting was 4 (0.7 nm) and the gain was set so that the energy reading is at mid scale. The conditions for determinations in the flame were as outlined in the Perkin-Elmer methods manual [ 51. Results and discussion In an attempt to measure zinc levels in various areas of fingernails directly with the graphite furnace, an anomalous behavior of the analyte was observed. With nail, hair, and cuticle, but not with blood, the appearance of a double peak was observed, Fig. 1. That both peaks were due to Zn and not to background absorption was established as follows: The use of the background corrector has little effect on the appearance of the two peaks. With or without the background corrector on, one observes the two peaks, Fig. 2. The use of the 2165 Cu line for nail sample indicated little background absorption, Fig. 3. The absorption during the atomization stage due to smoke using the 2165 line is minimal. The effect of varying atomization and charring temperatures on the formation of two peaks was studied. Increasing the charring temperature from 300” C to 1000°C at 100°C intervals demonstrated that the second peak gets smaller as the temperature increases, but even at lOOO”C, a trace of the second peak was
cuticle
.0320
Fig. 1. Recorder blood.
0156
mg
tracings
showing
Zn double
mg
peak
0477
formation
bla
mg
with nail. hair, and cuticle,
but not with
394
D2U
wire
,0196 mg
.0129mg Fig. 2. Recorder tracing of Zn determination
in nail with and without
II20 background
correction.
still clearly evident. Charring temperature above 4OO”C, however, &owed considerable loss of Zn. The atomization tem~eratu~ was also studied. Above 19OO~C considerable loss of Zn was noted. A comp~~son was made of nail determinations directly in the furnace with determinations of acid digest either in the furnace or in the flame. Acid digests exhibit only a single peak in the furnace. If one uses, with a solid sample, the peak area of both peaks for dete~in~ng Zn levels, the results agree fairly well with Zn levels determined on the acid digests, Table I. Solid sampling was employed on nail, samples from eleven individuals. Eight of these individu~s gave us enough material (same nail fragment) to prepare
2165
Cu line
2138
.0360 Fig. 3. Recorder
Zn
mg
tracing of check fox background
tine
.0320
mg
abslorption using the 2165 Cu line.
395
TABLE I ANALYSIS
FOR Zn IN FINGERNAILS:
Solid samples furnace Nitric acid digests furnace Nitric acid digests flame
COMPARISON
OF METHODS
@pm)
n
x
s
Range
11 8 7
108 163 163
k24.8 t30.7 t20.6
68-145 136-232 135-188
acid digests which were run both in the furnace and flame. The agreement between digests run in the furnace and flame is quite good. The agreement between values obtained by these methods and the direct determination in the furnace using the area of the two peaks obtained as a measure of Zn gives, in general, slightly lower results than the other techniques, but the data indicate that Zn is definitely coming off as two peaks. The appearance of absorption curves is influenced by a number of factors such as the form of the analytes, their volatility, the physical and chemical properties of the matrix, the nature of the furnace and the heating rate of the furnace. Several of these factors probably play a role in the observed behavior. The appearance of two discrete peaks due to zinc must be due to two discrete forms which volatilize and dissociate at clearly different rates. In an attempt to understand the anomalous behavior observed, the behavior of various zinc salts in non-aqueous media in the furnace was enlightening. Fig. 4 illustrates the behavior of ZnClz and ZnO in the furnace. ZnCl, gave a peak sooner than ZnO and a mixture of the two salts gave two peaks. Zinc chloride is known to boil at 732”C, and probably at this temperature dissociates fairly readily. Free Zn boils at 907°C. On the other hand, salts containing oxanions decompose to the oxide which is not markedly reduced to free gaseous zinc until about 1200°C. The metallic oxides in the graphite furnace are believed to be reduced to the free metal. In the case of ZnO, the following equation expresses the reaction. A zI%)
+ C(s) G
CO(,)
+ Z%g,
zno
ZllCI,
.
ZnClz 8
zno
84mm
mm --7o-
L
J
Fig. 4. Recorder tracing of modeling experiment iUustrating the behavior of different Zn salts.
396
TABLE
II
CONVERSION Time
OF
between
Hz0
and
ZnO
addition
introduction
IN AQUEOUS of
MEDIA Peak
into
TO
YIELD
height
84
mm
mm
74
mm
Peak
peak
74
PEAK height
mm
peak
furnace 30
s
62
2 min
50
119
3.5
min
80 mm
48
130
5 min
13
136
6.5
10
140
min
Gaseous dissociated Zn is not produced in this case until about 1200°C. Campbell and Ottaway [6] have presented evidence that carbon reduction leading to the direct liberation of gaseous metal ions plays an important role in the operation of the graphite furnace. The degree of contact of the analyte compounds with the walls of the furnace may partly explain the observed anomalous behavior. The influence of particle size was noted in that nail particles homogenized to an extremely fine powder tended to give only one Zn peak. Partical size may also influence the rate of heating of the sample. Further evidence that the intimacy of the analyte and the carbon of the furnace is a factor is presented in Table II. This data show8 the conversion in the presence of water of the slower oxide peak into the faster peak. Two peaks are observed 30 s after ZnO has been in contact with water. After 6 min contact with water, essentially a single peak is observed. TABLE
III
COMPARISON FURNACE
OF IN THE
Subject
DIRECT
DETERMINATION
DETERMINATION
OF
IN
FURNACE
Zn IN WHITE
WITH
SPOTS
AND
NITRIC
NORMAL
ACID NAIL
DIGESTS AREAS
Zn @pm) .___ Normal
area
n Direct
White x
n
s
spot
area x
s
determination f20.2
t23.8
B.A.
12
8
94.5
S.S.
15
114.9
+25.3
13
98.9
L.S.
9
145.5
+39.9
10
M.D.
5
100.0
r18.5
5
98.7
+9.3
L.T.
5
156.8
+32.1
8
107.1
+25.3
86.9
+27.2 k43.6
G.D.
8
67.6
8
64.5
i17.5
J.L.
8
89.2
i20.7
8
96.8
k12.1
P.P.
8
93.4
t27.5
8
100.3
i17.0
Nitric N.B.
acid
digests
?8.7
203.7
in furnace 4
74.5
t1.9
71.3
*2.1
4
84.0
+2.3
76.4
k1.8 t7.7
S.R.
4
152.1
k3.3
143.0
R.F.
4
61.4
+3.2
68.1
f8.6
4
61.1
t4.2
58.4
+2.3
4
69.2
k3.9
60.2
+4.1
4
81.7
24.2
93.2
t4.7
4
75.4
i2.6
106.7
k8.8
H
IN
397
TABLE IV COMPARISON
WITH DATA IN THE LITERATURE
Method
n
x
s
Range
AAS solid sampling AAS digest furnace AAS digest flame AAS Ill Emission [ 7 1 Neutron activation [8]
11 76 7 82 9 3
108 150 163 196 -
24.8 48.8 20.6 -
68-145 104-427 135-188 130-391 116-3080 1200-2700
The anomalous behavior observed with nails appears to be characteristic of Zn determinations in tissues having a low water content such as nail and hair. The water content of nails is less than l%, while that of hair is about 4%. Table III presents some representative data on the determination of Zn in white spots and normal nail areas. Several samples from a white spot were analysed and compared with nearby areas of normal nail. The samples from each individual came from the same nail. Evident is the great heterogeneity with regard to Zn levels of the minute area required for direct determination in the graphite furnace. The use of larger areas in the digests gives one an average value, but in either case, statistically significant differences in Zn levels between white spots and normal nail areas were not observed. Table IV compares our values with Zn values for fingernails reported in the literature. The determination of Zn in nails by solid sampling presents a number of difficulties. Because of the high sensitivity of the 2138 line only very small samples can be employed. The nail appears to be quite heterogeneous with regard to Zn levels. This heterogeneity in metal concentration has been previously reported for sodium, Kollberg and Ekbohm [9] and copper, Van Stekelenburg et al. [lo], Barnett and Kahn [ll]. Determination of Zn using the 3076 line was considered. However, the 3076 line is so much less sensitive that the sample size required must be so large that too much background interference was encountered. Although Zn levels can be determined on an individual nail fragment, the variation from fragment to fragment is such that it was possible to obtain statistically significant differences between white spots and normal nail areas. The use of digests in the furnace for both white spots and nails is probably preferred for determining Zn levels in nail in that the digests give an average value of a larger nail area, Zn is detected as a single peak. Zn determination of digests of nails agreed well with values reported in the literature. Our inability to demonstrate a significant difference in the Zn level of the white spot as compared with adjacent normal areas would indicate that any effects of Zn on white spot formation are indirect. It is conceivable that Zn may influence an enzyme system involved in the synthesis of nail tissue. The question of whether subjects who have white spots have lower Zn levels generally in their nails has been under study. Although we find a tendency in this direction, statistical significance is difficult to attain because of the hetero-
398
geneity of the samples involved. ther study.
The role of Zn in nail formation
requires fur-
References 1 2 3 4 5 6 7 8 9 10
Harrison, W.W. and Tyree, A.B. (1971) Clin. Chim. Acta 31, 63-73 Robson, J.R.K. and Brooks, G.J. (1974) Clin. Cbim. Acta 55, 255 Pfeiffer, C.C. and Jenney, E.H. (1974) J. Am. Med. Assoc. 228. 157 Muerhcke, R.C. (1956) Br. Med. J. 195, 1327-1328 Perkin-Elmer (1973) Analytical Methods for Atomic Absorption Spectrophotometry Campbell, W.C. and Ottaway, J.J. (1974) Talant 21, 873 Goldblu, R.W., Derbey, S. and Lerner, A.B. (1953) J. Invest. Dermatol. 20, 13 Petushkov. A.A.. Linekin, D.M., Balcius, J.F. and Brownell. G.L. (1969) Nuclear Med. 10, 730 Kollberg. H. and Ekbohm, G. (1974) Acta Pediatr. Stand. 63,405 Van Stekelenburg, G.J., van de Laar, A.J.B. and van der Lag, J. (1975) Clin. Chim. Acta 59, 233-
240 11 Barn&t, W.W. and Kahn, H.L. (1974)
Acta Pediatr. Sand.
63, 405
Clinica Chimica Acta, 70 (1976) 391-398 @ Elsevier Scientific Publishing Company,
Amsterdam
- Printed
in The Netherlands
CCA 7835
DETERMINATION OF ZINC IN FINGERNAILS ABSORPTION SPECTROSCOPY
ARTHUR
SOHLER,
PATRICK
WOLCOTT
Brain Bio Center, 1225 State Road,,Princeton, (Received
February
BY NON-FLAME
ATOMIC
and CARL C. PFEIFFER N.J. 08540
(U.S.A.)
11,1976)
Summary The determination of Zn in fingernails directly using the graphite furnace presented certain difficulties due to the anomalous behavior of the analyte in the furnace. The appearance of two peaks which were due to Zn and not to any background interference was noted. The Zn value obtained by adding the area of these two peaks compared fairly well with Zn levels determined by wet ashing and subsequent determination either in the furnace or flame. Wet ashed samples gave only a single peak. It was possible to produce a model of the phenomenon with various Zn salts in a non aqueous matrix. Under these conditions ZnS04 and ZnO gave a discretely different peak from ZnClz or metallic Zn. Several tissues such as serum, whole blood, cuticle, and hair were examined for multiple peak formation. Direct determination of Zn in fingernails with the graphite furnace is possible for certain applications such as the determination of Zn levels of white spots in fingernails. For this purpose it is possible to use a sample size as small as 20 pg using the 2138 Zn line. This allows one to run several determinations on a single white spot. However, where sample size is not a limitation, wet ash digestion prior to determination in the furnace is probably the preferred procedure.
Introduction The possibility of using trace metal analysis of readily available tissues such as nail or hair for diagnostic purposes has been considered by a number of investigators. Harrison and Tyree [l] have proposed the determination of Cu in nails for diagnosis of cystic fibrosis. Robson and Brooks [ 23 have demonstrated an increase in the sodium and calcium concentration of nails from children suffering from Kwashiorkor. The occurrence of white spots in the fingernails of many children, teenagers,
392
and a few adults and their possible ~lationship to a Zn deficiency was reported by Pfeiffer and Jenney [3]. Previously Muerhcke [4] had ascribed the phenomenon to a low serum albumin level. His largest group of patients suffered from nephrosis while the next most prevalent group suffered from hepatic cirrhosis. All had low serum albumin levels. A major portion of serum zinc, about 70% is bound to albumin so that zinc depletion could be responsible for the white spots. In addition, Pfeiffer reports that the white spots disappear in patients who are on zinc therapy. The present report is concerned with an attempt to determine whether there is a zinc deficiency in the white spots of nails per se or is the role of zinc in white spot formation more indirect. The use of flameless atomic absorption was necessitated by limitations of sample size and matrix and the sensitivities of available zinc lines. Two methods of sample introduction into the graphite furnace were compared. Because of the nature of the matrix, several problems were encountered which may have practical and theoretical implications with regard to flameless atomic absorption. Materials and methods Reagents and glassware Reagents were prepared and the final rinses of glassware were made with twice glass distilled water which was subsequently passed through a Barnstead research model mixed bed ion exchange resin. Baker analytical reagent grade nitric acid was used for the digestions This gave a lower blank for Zn than their Ultrex grade nitric acid. Standards were prepared by appropriate dilution of Zn standard 1000 ppm, Fisher certified for atomic absorption. Glassware for trace metal analysis was specially cleaned by soaking in 50% nitric acid for 30 min, subsequently rinsed with the double distilled, deionized water and dried in an oven. Collection and sample ~~~dl~~g Nail clippings were obtained free of gross cont~ination such as nail polish. Clippings were washed twice with acetone, and dried at 100°C for 10 min and allowed to cool in desiccator. White spots and adjacent normal nail areas were cut up into appropriate pieces for analysis, and weighed on a Sierra electromagnetic ultra microbalance. Acid digests of approximately 1 mg nail samples were prepared by adding 0.04 ml concentrated nitric acid in a 12 mm X 75 mm Pyrex test tube. The tube was covered with a glass marble and heated at 65°C for 1 h. Tubes were cooled and diluted with 0.96 ml of distilled deionized water. lo+1 samples of digests were introduced into the furnace using an Eppendorf micro pipetter. Solid nail samples (IO-20 pg) were weighed in tantalum boats on the ultra microbalance and introduced into the furnace with the Perkin-Elmer sampling spoon.
Apparatus A Perkin-Elmer
503 Atomic Absorption
Spectrophotometer
with background
393
corrector in conjunction with an HGA 2000 graphite furnace was employed in this study. A Perkin-Elmer Model 165 chart recorder was used to record the absorbance signal. The peak areas were determined by cutting out the peak and weighing it. Chart speed was usually 60 mm/min although occasionally 240 mm/min was used. The instrumental parameters were as follows: Zn was determined with the 2138 line using a Zn, Cu, Mn, Fe multielement lamp. Furnace parameters were: drying, 125”C, 30 s (seconds) ashing, 5OO”C, 30 s and atomization 1900°C for 10 s. Nitrogen flow rate was 5. On the spectrophotometer the slit setting was 4 (0.7 nm) and the gain was set so that the energy reading is at mid scale. The conditions for determinations in the flame were as outlined in the Perkin-Elmer methods manual [ 51. Results and discussion In an attempt to measure zinc levels in various areas of fingernails directly with the graphite furnace, an anomalous behavior of the analyte was observed. With nail, hair, and cuticle, but not with blood, the appearance of a double peak was observed, Fig. 1. That both peaks were due to Zn and not to background absorption was established as follows: The use of the background corrector has little effect on the appearance of the two peaks. With or without the background corrector on, one observes the two peaks, Fig. 2. The use of the 2165 Cu line for nail sample indicated little background absorption, Fig. 3. The absorption during the atomization stage due to smoke using the 2165 line is minimal. The effect of varying atomization and charring temperatures on the formation of two peaks was studied. Increasing the charring temperature from 300” C to 1000°C at 100°C intervals demonstrated that the second peak gets smaller as the temperature increases, but even at lOOO”C, a trace of the second peak was
cuticle
.0320
Fig. 1. Recorder blood.
0156
mg
tracings
showing
Zn double
mg
peak
0477
formation
bla
mg
with nail. hair, and cuticle,
but not with
394
D2U
wire
,0196 mg
.0129mg Fig. 2. Recorder tracing of Zn determination
in nail with and without
II20 background
correction.
still clearly evident. Charring temperature above 4OO”C, however, &owed considerable loss of Zn. The atomization tem~eratu~ was also studied. Above 19OO~C considerable loss of Zn was noted. A comp~~son was made of nail determinations directly in the furnace with determinations of acid digest either in the furnace or in the flame. Acid digests exhibit only a single peak in the furnace. If one uses, with a solid sample, the peak area of both peaks for dete~in~ng Zn levels, the results agree fairly well with Zn levels determined on the acid digests, Table I. Solid sampling was employed on nail, samples from eleven individuals. Eight of these individu~s gave us enough material (same nail fragment) to prepare
2165
Cu line
2138
.0360 Fig. 3. Recorder
Zn
mg
tracing of check fox background
tine
.0320
mg
abslorption using the 2165 Cu line.
395
TABLE I ANALYSIS
FOR Zn IN FINGERNAILS:
Solid samples furnace Nitric acid digests furnace Nitric acid digests flame
COMPARISON
OF METHODS
@pm)
n
x
s
Range
11 8 7
108 163 163
k24.8 t30.7 t20.6
68-145 136-232 135-188
acid digests which were run both in the furnace and flame. The agreement between digests run in the furnace and flame is quite good. The agreement between values obtained by these methods and the direct determination in the furnace using the area of the two peaks obtained as a measure of Zn gives, in general, slightly lower results than the other techniques, but the data indicate that Zn is definitely coming off as two peaks. The appearance of absorption curves is influenced by a number of factors such as the form of the analytes, their volatility, the physical and chemical properties of the matrix, the nature of the furnace and the heating rate of the furnace. Several of these factors probably play a role in the observed behavior. The appearance of two discrete peaks due to zinc must be due to two discrete forms which volatilize and dissociate at clearly different rates. In an attempt to understand the anomalous behavior observed, the behavior of various zinc salts in non-aqueous media in the furnace was enlightening. Fig. 4 illustrates the behavior of ZnClz and ZnO in the furnace. ZnCl, gave a peak sooner than ZnO and a mixture of the two salts gave two peaks. Zinc chloride is known to boil at 732”C, and probably at this temperature dissociates fairly readily. Free Zn boils at 907°C. On the other hand, salts containing oxanions decompose to the oxide which is not markedly reduced to free gaseous zinc until about 1200°C. The metallic oxides in the graphite furnace are believed to be reduced to the free metal. In the case of ZnO, the following equation expresses the reaction. A zI%)
+ C(s) G
CO(,)
+ Z%g,
zno
ZllCI,
.
ZnClz 8
zno
84mm
mm --7o-
L
J
Fig. 4. Recorder tracing of modeling experiment iUustrating the behavior of different Zn salts.
396
TABLE
II
CONVERSION Time
OF
between
Hz0
and
ZnO
addition
introduction
IN AQUEOUS of
MEDIA Peak
into
TO
YIELD
height
84
mm
mm
74
mm
Peak
peak
74
PEAK height
mm
peak
furnace 30
s
62
2 min
50
119
3.5
min
80 mm
48
130
5 min
13
136
6.5
10
140
min
Gaseous dissociated Zn is not produced in this case until about 1200°C. Campbell and Ottaway [6] have presented evidence that carbon reduction leading to the direct liberation of gaseous metal ions plays an important role in the operation of the graphite furnace. The degree of contact of the analyte compounds with the walls of the furnace may partly explain the observed anomalous behavior. The influence of particle size was noted in that nail particles homogenized to an extremely fine powder tended to give only one Zn peak. Partical size may also influence the rate of heating of the sample. Further evidence that the intimacy of the analyte and the carbon of the furnace is a factor is presented in Table II. This data show8 the conversion in the presence of water of the slower oxide peak into the faster peak. Two peaks are observed 30 s after ZnO has been in contact with water. After 6 min contact with water, essentially a single peak is observed. TABLE
III
COMPARISON FURNACE
OF IN THE
Subject
DIRECT
DETERMINATION
DETERMINATION
OF
IN
FURNACE
Zn IN WHITE
WITH
SPOTS
AND
NITRIC
NORMAL
ACID NAIL
DIGESTS AREAS
Zn @pm) .___ Normal
area
n Direct
White x
n
s
spot
area x
s
determination f20.2
t23.8
B.A.
12
8
94.5
S.S.
15
114.9
+25.3
13
98.9
L.S.
9
145.5
+39.9
10
M.D.
5
100.0
r18.5
5
98.7
+9.3
L.T.
5
156.8
+32.1
8
107.1
+25.3
86.9
+27.2 k43.6
G.D.
8
67.6
8
64.5
i17.5
J.L.
8
89.2
i20.7
8
96.8
k12.1
P.P.
8
93.4
t27.5
8
100.3
i17.0
Nitric N.B.
acid
digests
?8.7
203.7
in furnace 4
74.5
t1.9
71.3
*2.1
4
84.0
+2.3
76.4
k1.8 t7.7
S.R.
4
152.1
k3.3
143.0
R.F.
4
61.4
+3.2
68.1
f8.6
4
61.1
t4.2
58.4
+2.3
4
69.2
k3.9
60.2
+4.1
4
81.7
24.2
93.2
t4.7
4
75.4
i2.6
106.7
k8.8
H
IN
397
TABLE IV COMPARISON
WITH DATA IN THE LITERATURE
Method
n
x
s
Range
AAS solid sampling AAS digest furnace AAS digest flame AAS Ill Emission [ 7 1 Neutron activation [8]
11 76 7 82 9 3
108 150 163 196 -
24.8 48.8 20.6 -
68-145 104-427 135-188 130-391 116-3080 1200-2700
The anomalous behavior observed with nails appears to be characteristic of Zn determinations in tissues having a low water content such as nail and hair. The water content of nails is less than l%, while that of hair is about 4%. Table III presents some representative data on the determination of Zn in white spots and normal nail areas. Several samples from a white spot were analysed and compared with nearby areas of normal nail. The samples from each individual came from the same nail. Evident is the great heterogeneity with regard to Zn levels of the minute area required for direct determination in the graphite furnace. The use of larger areas in the digests gives one an average value, but in either case, statistically significant differences in Zn levels between white spots and normal nail areas were not observed. Table IV compares our values with Zn values for fingernails reported in the literature. The determination of Zn in nails by solid sampling presents a number of difficulties. Because of the high sensitivity of the 2138 line only very small samples can be employed. The nail appears to be quite heterogeneous with regard to Zn levels. This heterogeneity in metal concentration has been previously reported for sodium, Kollberg and Ekbohm [9] and copper, Van Stekelenburg et al. [lo], Barnett and Kahn [ll]. Determination of Zn using the 3076 line was considered. However, the 3076 line is so much less sensitive that the sample size required must be so large that too much background interference was encountered. Although Zn levels can be determined on an individual nail fragment, the variation from fragment to fragment is such that it was possible to obtain statistically significant differences between white spots and normal nail areas. The use of digests in the furnace for both white spots and nails is probably preferred for determining Zn levels in nail in that the digests give an average value of a larger nail area, Zn is detected as a single peak. Zn determination of digests of nails agreed well with values reported in the literature. Our inability to demonstrate a significant difference in the Zn level of the white spot as compared with adjacent normal areas would indicate that any effects of Zn on white spot formation are indirect. It is conceivable that Zn may influence an enzyme system involved in the synthesis of nail tissue. The question of whether subjects who have white spots have lower Zn levels generally in their nails has been under study. Although we find a tendency in this direction, statistical significance is difficult to attain because of the hetero-
398
geneity of the samples involved. ther study.
The role of Zn in nail formation
requires fur-
References 1 2 3 4 5 6 7 8 9 10
Harrison, W.W. and Tyree, A.B. (1971) Clin. Chim. Acta 31, 63-73 Robson, J.R.K. and Brooks, G.J. (1974) Clin. Cbim. Acta 55, 255 Pfeiffer, C.C. and Jenney, E.H. (1974) J. Am. Med. Assoc. 228. 157 Muerhcke, R.C. (1956) Br. Med. J. 195, 1327-1328 Perkin-Elmer (1973) Analytical Methods for Atomic Absorption Spectrophotometry Campbell, W.C. and Ottaway, J.J. (1974) Talant 21, 873 Goldblu, R.W., Derbey, S. and Lerner, A.B. (1953) J. Invest. Dermatol. 20, 13 Petushkov. A.A.. Linekin, D.M., Balcius, J.F. and Brownell. G.L. (1969) Nuclear Med. 10, 730 Kollberg. H. and Ekbohm, G. (1974) Acta Pediatr. Stand. 63,405 Van Stekelenburg, G.J., van de Laar, A.J.B. and van der Lag, J. (1975) Clin. Chim. Acta 59, 233-
240 11 Barn&t, W.W. and Kahn, H.L. (1974)
Acta Pediatr. Sand.
63, 405
