-------
CVI
O
O
in
CD
Q.
JD
CO
ru
I I
00 ID
rH rH
• t
G) G)
00 10
G) G)
cs
CO
ID
GO
T3
U
ra
D-
i
D-
J—
II
D-
D-
ID
CD
•
C5>
kS
fXJ
U 0
I I
ru cs
O-O.
G) G) G) G) CO CS3
l HH Oi UJ O f—
-------
\
ro
in
n
I
rxj
•
t-H
O
1!
CD
CL
CO
in
I
in
_JO CD 0-0-
Ln
en
•
CS
OK-
in
CD
LJJ o i
Ln
r-
•
cs
- OJ
tn
n
in
i—i
•
r-J
. Ln
en
•
cs>
•a
.O
D_
i
0-
I—
11
O-
DL.
CD
CD
Ln
- co
- r-
i
LD
CS)
Ln
ID
•
CS)
o -o
•P O
3 ®
O ft
•H
-P O
^ X
03 -P
ft
h
w o
O «H
n Ou
n
(0 3
c
O O
n o
-------
I •
in
*H
•
G3
LO
o
It
CD
Q_
CO
h-QO.
I •
CD
CD
II
in
CD
•
CD
tn
*H
•
. CD
CD
O
UJ
C3
II
O-
ca
in
CD
cs
CO
•
C9
CD
- CD
CD
m .
> to
co
^o»
CO T-l
•H
X -
0 M
0)
«
0-1
> t
r- 1 flj
O C
W 3
0) >o
o
tO
o > e
tH
c o c
o en -H
o] n
•H ^i c
V< T3 O
«J -H
Cu ® -P
£ -P «
O OJ fc
O «H -P
r-H C
3 <»
• « O
CD I* C
. © O
CQ
-------
APPENDIX C
ANALYTICAL PROCEDURES USED IN THIS STUDY
Taken from CBL's Nutrient Analytical Services Laboratory Methods Booh
1. ORTHOPHOSPHATE
2. NITRITE+NITRATE
3. TOTAL DISSOLVED N AND P
4. TOTAL P (ACID PERSULFATE)
5. KJELDAHL N
6. PARTICULATE N (DIRECT)
7. PARTICULATE P (DIRECT)
-------
Nutrient Analytical Services Laboratory
STANDARD OPERATING PROCEDURES
Christopher F. D'Elia
Nancy L. Kaumeyer
Carolyn L. Keefe
Diane L Shaw
Kathryn V. Wood
Carl F. Zimmermann
Chesapeake Biological Laboratory (CBL)
University of Maryland
Box 38
Solomons, Maryland 20688
April 1987
-------
Nitrate ± Nitrite;
Filtered samples are passed through a granulated copper cadmium
column to reduce nitrate to nitrite. The nitrite (originally
present plus reduced nitrate) is then determined by diazotizing
sulfanilamide and coupling with N-1-napthylethylenediamine
dihydrochloride to form a colored azo dye. Nitrate is obtained
subtracting NC>2+ NC^ from N02 values.
Methodology: Technicon Industrial Method: 158-71 W/A
EPA. 1979. Chemical Analysis of Water and Waste
USEPA-600/4-79-020. Method §353.2.
Manifold Assembly ; See figure ?.
fLtarLdard Calibration Settings:
Yellow/Orange Sample Tubes: 2.0, 1.0, 0
Black/Black Sample Tubes: 9.0, 6.0, 2.0
Hamp; Normal
Sampling £a£e,: 40 hours 9:1 sample/wash ratio
550 nm
PJieiP_fciib_e.: 199-B021-01 Flowcell; 50 mm
lnt_ej:-fej: .ericas : Metal ions may produce a positive error if pres
in sufficient concentrations. The presence of
large concentrations of sulfide and/or sulfate
will cause a large loss of sensitivity to the
copper-cadmium column.
Reagents;
1. Ammonium £hlo_rlde. Reagent ;
Ammonium Chloride (NH4C1) 10.0 g
Alkaline Water 1000 ml
Dissolve 10.0 g (KH^Cl) in alkaline water and dilute
one liter. Alkaline water is prepared by adding " 2
concentrated Ammonium hydroxide to one liter of deion
water. Should attain a pH balance of 8.5.
2. £o_lcj: Reagent;
Sulfanilamide (CgHoN202S) 20.0 g
Concentrated Phosphoric Acid (F^PC^) 200.0 ml
N-1-naphthylethylenediamine dihydrochloride
(C12H14N2 * 2HC1) 1«° 9
Deionized Water 2000 ml
Brij-35 1.0 ml
-------
To approximately 1500 ml of deionized water, carefull
200 ml of concentration l^PC^ and 20 g of sulfanilami
Dissolve completely (heat necessary). Add 1.0 g of N
naphthylethylenediamine dihydrochloride and dissolve.
Dilute to 2 liters with deionized water and add 1.0 m
Brij-35. Store in a cold, dark place.
Preparation oj[ cr>ppe_r-j:adiiiium column;
1. Use good quality cadmium filings (25-60 mesh size).
2. Ten grams of cadmium are cleaned with 50 ml of 6 N HC1
one minute. Decant the HC1 and wash the cadmium with
another 50 ml portion of 6 N HC1 for one minute.
3. Decant the HC1 and wash the cadmium several times with
distilled water.
4. Decant the distilled water and add 50 ml of 2% (W/V) C
* 5HoO. Wash the cadmium until no blue color remains
solution.
5. Add another 50 ml of 2 CuS04 * 5H2O and wash the cadmi
until no blue color remains in solution.
6. Decant and wash throughly (approximately 10 times) wit
deionized water..
7. Fill the reductor column with ammonium chloride reagen
transfer the prepared cadmium particles to the column
a Pasteur pipette. lie. careful not to &llo_w_ any sJji
t>ui>bJLe_£. io, b_£ trapped In i_h£ cjiliinn*. The column is a
cm length of 0.110" ID tubing.
8. When the entire column is fairly well packed with grar
insert glass wool plugs at both ends of the column, wi
reagents running through the system attach the column.
£e_JQe.m.b_e_L to have no air bubbles in the valve and to at
the column to the intake side of the valve first.
9. Check for good flow characteristics (good bubble patt<
If the column is packed too tightly, you will get an
inconsistent flow pattern will result.
Prior to sample analysis, condition the column with approximate
100 mg N (nitrate)/! for 5 minutes followed by 100 mg N (nitrite)/!
10 minutes.
-------
Standards
A. £iL££k Standard; Dissolve 0.5055 g KNC^ into one liter
deionized water (1 ml = 5 ug-at N).
B. ttarklng Standard &: 0.8 mis of stock standard up to II
with deionized water yields 40 ug at N/l (0.56 mgN/1).
C. Harking Standard; 0.8 mis of stock standard up to 200
with deionized water yields 20 ug at N/l (0.28 mgN/1).
1.0 mis and 1.5 mis of stock standard up to 100 ml wit
deionized water yields 50 and 75 ug at N/l, respective
(.70 and 1.05 mg N/l) for use with the orange-yellow s;
tube and yellow-blue NI^Cl tube employed with sample
concentrations < 0.56 mg N/l (N03~ + N02~).
2.5, 5.0, 10.0, 15.0, 25.0 mis of working standard A u
100 ml with deionized water yields 1.0, 2.0, 4.0, 6.0 i
10.0 ug at N/l or .014, .028, .056, .084, and .14 mg N
respectively.
-------
Cadmium
Redactor
Tube
MANIFOLD COE'IGURATION FOR NITRATE
To Sampler Wash Receptacle-
A2
5 Turns
22 Turns
Debubbler
Waste —
Waste
COLORIMETER
550 ran
50 mm F/C x 1.5 mm ID
199-B021-01 Phototube
GRN/GRN (Water)
BLK/BLK (Air)
(Ammonium Chloride)
BLK/BLK (Sample)
BLK/BLK (Air)
BLK/BLK (Color Reagent)
WHT/WHT-
(From. F/C)
Note: If sample concentration >.56 mgN/1
substitute:YEL/BLU for Ammonium Chloride
ORN/YEL for Sample
-------
Orthophosphate :
Ammonium molybdate and antimony potassium tartrate react in an acid
medium to form an antimony — phosphomolybdate complex which is
reduced to an intensely blue colored complex by ascorbic acid.
Methodology; Technicon Industrial Method No. 155-71W
EPA. 1979. USEPA-600/4-79-020. Method §365.1
Assembly: See figure ?.
Standard Calibration Settings; 9.0, 6.0, 3.0
fiamp; Normal
.Sampling. Eat£; 40/hr. 9:1 sample/wash ratio
£iltej:: 880 nm
Phototube; 19S-B021-04 Flowcell; 50 mm
Jnterferencesi Silicon at a level of 100 ug at Si/1 causes an
interference equivalent to approximately 0.04 ug
at P/l.
Reagents;'
1. Sulfuric Acid (4.9N);
Sulfuric Acid (H^SC^), concentrated
(sp. gr. 1.84) 136 ml
Deionized Water (QS to ) 1000 ml
Add 136 ml cone. H2S04 to approximately 800 ml good
quality deionized water while cooling (cold water
bath). After the solution is cooled, dilute to one
liter with deionized water.
2.
Ammonium Molybdate [(NH^g Mc-yC^ * 4 H2] 40 g
Deionized Water * 1000 ml
Dissolve 40 g of ammonium molybdate in 800 rnl of
deionized water. Dilute to one liter with deionized
water. Store in plastic bottle away from direct
sunlight.
3. Ascorbic; Acid.:
Ascorbic Acid (CgH8Og) 18.0 g
Deionized Water 1000 ml
-------
Dissolve 18 g. of ascorbic acid in 800 ml. deionized water.
Dilute to one liter with deionized water and dispense
(approx.) 40 ml. into clean polybottles and freeze.
4. Antimony Potassium Tartrate.:
Antimony Potassium Tartrate [(K(SbO)C4H4Og * 1/2 H23 3.0 g
Deionized Water 1000 ml
Dissolve 3.0 g antimony potassium tartrate in 800 ml
deionized water. Dilute to one liter with deionized
water .
5. .Sodium Lauiyl Sulfate. JSLS1:
Sodium Lauryl Sulfate (Sodium Dodecyl Sulfate M.W. =
288.38; Phosphate _£ 0.0001%) 3.0 g
Deionized water 100 ml
Dissolve 3.0 g SLS in 80 ml deionized Water.
Dilute to 100 ml with deionized water.
6. J&rkins. Reagents:
a. Reagent A: Sulfuric Acid (4.9N) 50 ml
Ammonium Molybdate 15 ml
Antimony Potassium Tartrate 5 ml + 1 ml SLS
b. Reagent B: Ascorbic Acid 30 ml + 0.3 ml SLS
Standards
A. .Stasis. StanojLld.: Dissolve 1.632 g KtUPCU into one
liter deionized water and add 1.0 ml chloroform as
a preservative (1 ml = 12 ug at P).
B. Secondary standard; Take 1.0 ml of stock standard
and dilute to 100 ml with deionized water (0.12)ug
at P/ml).
C. EfilKina ^iand^rds.: 0.1, 0.25, 0.5, 2.5 and 5 mis of
B up to 100 ml with deionized water yield
concentrations of 0.12 ug at/1 (0.00372 rog/1),
0.3 ug at/1 (0.0093 mg/1), 0.6 ug at/1 (0.0186 mg/1),
1.2 ug at/1 (0.0372 mg/1), 3.0 ug at/1 (0.093 mg/1)
and 6.0 ug at/1 (0.186 mg/1).
-------
Organic Analytes
Bationale:
Dissolved organic Carbon, Nitrogen and Phosphorus are described
below. All procedures except Kjeldahl require the addition of potassium
persulfate to a sample, which when under heat and pressure break down the
organic constituents to inorganic forms. Inorganic fractions are then
subtracted from the total dissolved sample to yield the dissolved organic
concentration. (Figure _ and _ ) .
Sampling .and S±cjr.ase_:
Surface, bottom, above and below pycnocline water samples are
collected via a submersible pump system. Collected water samples are
filtered through GF/F filters (nominal pore size 0.7 um) and placed in
appropriate containers and preserved (Table _ ) .
Analyte Volume .Storage
Dissolved Organic Carbon "20 Freeze
Dissolved Nitrogen/Phosphorus 10 Freeze
Dissolved Phosphorus (Acid Persulfate) 20 Freeze
Dissolved Kjeldahl "50 JSO
Total Dissolved Nitrogen and. Phosphorus:
The method utilized is that of D'Elia, et al. 1977. This method is
a persulfate oxidation technique for nitrogen and phosphorus where,
under alkaline conditions, nitrate is the sole N product and
phosphate is the sole P product.
Assembly.: Same as nitrate and phosphate.
Normal
Eatej. 40/hr 9:1 sample/wash ratio
550 run for nitrate; 880 nm for orthophosphate
: 199-B021-01 for nitrate; 199-B021-04 for orthophosphate
Flowcells; 50 mm
Int£r.£ejL£n£es_: Metal ions may produce a positive nitrate error if
present in sufficient concentrations. The presence
of large concentrations of sulfate will cause a
large loss of sensitivity to the copper-cadmium
column. Silicon at a level of 100 ug at Si/1 causes
an interference equivalent to approximately 0.04 ug
al P/l.
-------
Outline.
1. Ten mis of filtered water (GF/F, 0.7 urn) is placed in a 30
ml screw cap test tube and frozen.
2. When ready to analyze, thaw samples and bring to room
temperature.
3. Add 15.0 ml oxiding reagent (Mg(OH)2). A precipitate will
form with seawater samples. Test tubes are capped fairly
tightly.
4. Samples are then autoclaved at 100-110OC (between 3-4 psi)
for 30 minutes and slowly brought back to atmospheric
pressure.
5. Tubes are removed and cooled to room temperature (samples
can be stored at this point) .
6. Add 1.5 ml 0.3N HC1 to each cample.
7. Mix with Vortex mixer until precipitate dissolves.
8. Add 2.0 ml buffer solution to each tube. The pH of the
sample should be 7-8 after the addition of the buffer
•solution.
9. Analyze for NC>2~ + ^3" and P0^~ (see dissolved inorganic
section) .
Reagentg
1. Butfejc. solution!
30.9 g HoBC>3 (Boric Acid) dissolved in approximately 800 ml
deionized water. Add 101 ml of a 1M NaOH solution (40 g
NaOH/1) to the H3B03 solution and bring up to one liter
with deionized water. The solution is stable for many
weeks .
2. 0.3N UCli
2.5 ml concentrated HC1 brought up to 100 ml with deionized
water .
3. Oxidizing. Eeassnt:
3.0 g NaOH and 6.7 g of low N ( <0.001%) potassium
persulfate (K2S2Og) are dissolved in one liter of deionized
water just before use.
-------
1. The use of internal organic standards (glutamic acid and
glycerophosphate) allows to check for percent recovery and
is routinely used at CBL.
2. The procedure includes an internal dilution factor of
samples and standards due to addition of reatgents of 2.85.
3. Reagent Blanks: Reagents only are digested in 30 ml test
tubes, neutralized and buffered. The analyzed peak heights
of lO-j and PC>4 are normalized to the sample + reagent
volume by multiplying by 18.5/28.5. The re£;ultant
normalized reagent blank peak height is then subtracted
from the sample peak heights before calculating the
concentrations based on the peak heights of the standards.
Preparation of Internal standards:
A. Stock Glutamic Acid Standard.: Dissolve .3705 g glutamic
acid in approximately 400 ml deionized water and then bring
up to 500 ml with deionized water. Add 03 ml chloroform
to act as a preservative.
B. Eorking. Glutamic. Mid. Standard; 1 ml of A up to 100 mis
with deionized water will yield 50.4 ug at 1-1/1 (0.7056 mg
N/l).
C. Stock. Gly££I2EbQSpJ3at£ .Standard; Dissolve 0.0473 g B-
Glycerophosphoric Acid, Disodium Salt, 5-Hyc3rate in
approximately 400 ml deionized waterand then bring up to
500 ml with deionized water. Add 0.5 ml chloroform to act
as a preservative.
D. Hoxking. filycersEhfigpliate Standard.: 1 ml of c up to 100 mis
with deionized water will yield 3.09 ug at E/l (0.096 mg
P/l).
Preparation. of Ksrk-ing. inorganic, standards:
A. Stock Mtrate. Standard; From nitrate method.
B. Eorking. nitrate standards; 0.5, i.o and 1.5 ml of Nitrate
Stock Standard A up to 100 ml with deionized water will
yield 25 ug at N/l (.35 mg N/l), 50 ug at N/l (.70 mg N/l)
and 75 ug at N/l (1.05 mg N/l), respectively.
C. Stock OrthopJaoSEnatfi Standard; From othophosphate method.
D. Secondary. OrthOpJasspJaate. Standard; From othophosphate
method.
-------
E. Korklng d±b2pJ3aspJ3at£ Standards-; 0.5, 1.0 and 2.5 mis of
Secondary Orthophosphate Standard D up to 100 ml with
deionized water will yield 0.6 ug at P/l (.0186 mg P/l),
1.2 ug at P/l (.0372 mg P/l) and 3.0 ug at P/l (.093 mg
P/l), respectively.
Phosphorus (Acid
The method used by CBL personnel is that of Menzel, D.W. and N.
Corwin (1S65) .
Outline.
1. Prepare 0-5% solution of K2S2Og.
a. 25g IvjS^Og up to 500 mis with deionized water.
b. 12.5 g Y^lPs UP tc 250 mis with deionized water.
2. To each 20 ml of sample (in 30 ml screw cap test-tube) add 3.2
ml of the 5% I^^Og solution and shake.
3. Place tubes in pressure cooker at 3-4 psi for one hour.
4. 20 mis of standards (3 replicates) are placed in 30 ml test-
tube and treated in exactly the same manner as the samples.
5. Blanks (3 replicates) consist of 20 ml deionized water and
then treated in exactly the same manner as the samples.
6. Aliquot of cooled, shaken sample transferred to AutoAnalyzer
cup with Pasteur pipette.
7. Phosphate analyzed.
Menzel, D.W. and N. Corwin. 1965. The measurement of
total phosphorus in seawater based on the liberation
of organically bound fractions by persulfate
oxidation. UjmoL. <2££gncat*- 10:280-282.
£Sanifold. Msenbly: See figure 1
I&rjD: Normal
Sampling Eat£: 40/hr 9:1 SampleA7ash Ratio
880 nm
199-B021-04
-------
Reagents;
1. Eeifinized Ifeter Diluent;
Add .5 g sodium lauryl sufate (SLS) to 500 ml good quality
deionized water. Mix well!
2. JlsUlfujQc; Ac_id:
From orthophosphate method.
3. AjmeniuiB Molybdate;
From orthophosphate method.
4. Ascorbic, A£id:
From orthophosphate method.
5. Mfcimcny pj?ias£ium Tartrate;
From orthophosphate method.
6. Sediym Laurel Sulfate.
From orthophosphate method.
7. iJoiMng. Eesgents.:
From orthophosphate method.
Standards.;
A. Stock Standard; KH2P04; from Orthophosphate method.
B. Secondary. Standard.: from Orthophosphate method.
C. Working Standards; Take 0.5, 1.0, 2.0, 2.5 ml of
Secondary Standard E and dilute each to 100 ml with
deionized water which will yield 0.6 ug at P/l (.0186 mg
P/l); 1.2 ug at P/l (.0372 mg P/l); 2.4 ug at P.I (.0744 mg
P/l) and 3.0 ug at P/l (.093 mg P/l).
D. £tQ£k. i^c^jr^Ehosphaie. Standard.: From alkaline persulfate
method.
E. Hctking. filyi^spJb^p-hate. Standard; Take i.o ml of stock
Glycerophosphate Standard B and dilute to 100 ml with
deionized water which will yield 3.09 ug at P/l (.096 mg
P/l).
-------
Manifold Configuration for Total Phosphorus
(Acid Persulfate)
To sampler wash receptacle
37°C 5 turns
Heating
Bath
5 turns
ooioo
Colorimeter
880 nm filters
50 x 1.5 flow cell
199-B021-04 Phototubes
5 turns
IPPQO
Waste
GRN/GRN (water)
BLK/BLK (air)
RED/RED (deionized water)
ORN/ORN (sample) 1
Sampler
40/hr
9:1
ORN/WHT (Reagent A)
ORN/GRN (Reagent B)
WHT/WHT (From F/C)
-------
MANIFOLD CCNFIGURATIC3N FOR PBQSPBATE
37°C
Heating
Bath
To Sampler Wash Receptacle-]
5 Turn
COLORIMETER
880 run filters
50 mm F/C x 1.5 mm ID
199-B021-04 Phototube
5 Turns
Waste
GRN/GRN (Water)
BLK/BLK (Air)
(Sample)
ORN/WHT (Reagent A)
ORN/GRN (Reagent B)
WOT/WHT (From F/C)
Sampler
40/hr.
9:1
-------
Nitroen:
The sample is heated with a teflon boiling ball in the presence of
sulfuric acid/ potassium sulfate and mercuric sulfate for 3.5 hours.
The residue is cooled/ diluted to the original volume and is then
analyzed for ammonium. The ammonium determination is based on a
colormetric method in which an emerald-greem color is formed by the
reaction of ammonia with sodium salicylate, sodium nitroprusside and
sodium hypochlorite in a buffered alkaline medium at a pH of 12.8-
13.0. The ammonia salicylate complex is read at 660 nm using an
automated analyzer.
Digestion
Reagents;
1. stock Msisuiic. .Sulfate.:
Mercuric Oxide, Red (HgOO 8 g
Sulfuric Acid/ (H2S04) ; concentrated 10 ml
Diluted to 100 ml with ammonia free deionized water.
2. Digestion
Potassium Sulfate (K2S04) 135 g
Sulfuric Acid (concentrated) 200 ml
Stock Mercuric Sulfate 25 ml
Distilled Water qs 1000 ml
Dissolve 135 g of K^SOx in approximately 500 ml deionized
water and slowly, add 200 ml concentrated H^SO^ Add 25 ml
mercuric sulfate solution, let cool and dilute to 1000 ml
with deionized water.
Digestion
1. A 25 ml sample is added to each digestion tube.
2. Five (5 ml) of digestion solution and two teflon boiling
balls (Fisher Scientific) are then addded to each tube and
mixed with a vortex mixer.
3. SILICONE AIRTIGHT PLUGS ARE INSERTED IN THE DIGESTION TUBE
WHENEVER THEY ARE NOT BEING HEATED.
4. The digestion tubes are then heated in a block digestion
at 200°C for 1 hour and then at 360°C for 2.5 hours.
5. The tubes are then taken off the digestion and allowed to
cool for 15 minutes. Approximately 15 mis of deionized
water are then added to each tube (to dissolve any
precipitate) and capped. Allow to stand overnight.
-------
6. The following day, bring up to 25 ml volume with deionized
water (digestion tubes have been pre-marked) .
Cleaning Digestion Tubes; 25 mis of deionized water are added to each
tube and boiled at 200°C until dry. You may need to rinse the tubes with
20% NaOH followed by numerous deionized water rinses.
Analysis.
BeasentS.:
A. Sulfillic, Acid. £anpl£r_ i&gh ^slutiouj.
Potassium Sulfate (K2S04) 34 g
Sulfuric Acid 50 ml
Deionized water up to 1 ml
To approximately 800 ml deionized water acid 34 g KoSC^ and
dissolve. Slowly add 50 ml concentrated 112804 an dilute
to 1 liter with deionized water.
B. sodium chloride. Diluent Solution:
Sodium Chloride 10 g
Deionized water qs 1000 ml
C. "Sodium Byudiaxidfi Solution:
Sodium Hydroxide 200 g
Deionized water qs 1000 ml
To approximately 600 ml deionized water CAREFUTiTiY and
SLOWLY add 200 g NaOH. Please wear goggles! A great deal
of heat will be liberated. After the solution has cooled,
dilute to 1 liter with deionized water.
D. 5odiiffli f&licylai^j/s^djjjm iiitiopjriissid^ Solution:
Sodium Salicylate 70.0 g
Sodium Nitroprusside 0.3 g
Deionized water qs 1000 ml
BRLJ - 35 1 ml
E. Sodium Byj30£niacis3£ Solution:
Sodium Hypochlorite (Clorox) 12 ml
Deionized waer qs 200 ml
F. stock Buffer. Solution:
Sodium Phosphate, dibasic (Ka2 HP04 7H20) 134 g
Sodium Hydroxide 20 g
Deionized water qs 1000 ml
-------
Heat to dissolve 134.0 g of sodium phosphate, dibasic
H PO^) in approximately 800 ml deionized water. Add 20.0 g
of sodium hydroxide and dilute to 1 liter.
G.
* Sodium Potassium Tartrate 50 g
Stock Buffer solution 200 ml
NaOH solution (20% w/v) 100 ml
BRIJ 0.3 ml
Deionized water qs 1000 ml
* Fifty (50) grams of Sodium Potassium tartrate is added
to approximately 600 ml deionized water. (This is added as
a solid to avoid the rapid formation of mold during storage
of a 20% w/v Sodium Potassium Tartrate Stock Solution.)
200 ml of Stock buffer, 100 ml of sodium hydroxide solution
are then added. Deionized water is used to dilute to 1
liter and 0.3 ml BRLJ is added as the wetting agent.
Analysis. Procedure;
1) With the system pumping and deionized water flowing through the
system, add all the reagent lines EXCEPT the Salicylate/
Nitroprusside Line. After approximately ten minutes, add the
Salicylate/Nitroprusside line. If the pH of the flow stream is
low, the sodium salicylate reagent will precipitate.
2) Prepare standards and blanks in exactly the same manner as
samples — taking them all through the digestion procedure.
-------
c
a>
en
o
s-
(O
T3
(1)
i-
O
c
o
(O
3
01
O
o
(O
+J
3
O
VI
to
«
S-
QJ
oo
o
00
CM
3C
en
z
cc
u
c
$— °
3°
•^s
LO°
O
u
CD
c
•f—
4->
fO
OJ
J= CO
•••••
y
3
4->
O
O
.C
C3.
r^
O
1
1^
c^o
o
CO
1
CT>
CD
'—
to
OJ
3
O
-M
O
Q.
^^
1~ 0
O 1
r—
CSJ
O
CO
1
en
en
•—
-------
Particulate Analtes
The direct measurement of particulate C, N & P is the preferred
method used in this laboratory. It is felt that the greater volume
filtered onto the pad yields a more representative sample. The
alternative, subtraction of the dissolved from the total sample to
determine the particulate concentration often yields negative values
is totally unacceptable. Direct mesurement is rapid, more sensitive
more precise.
ajjd. Storage:
Surface, bottom, above and below pycnocline water samples are
collected via a submersible pump system. A known volume of the coll
water is filtered through GF/F filters (nominal pore size 0.7 urn), t
filter folded, placed in aluminum foil and frozen until analysis.
Particulate Carbon .and, Particulate Nitrogen Analysj-Si
Outline
1. A known volume of water is filtered onto a 25 mm precombust
GF/F (nominal pore size 0.7 urn) filter pad.
2. Duplicate sample taken
3. Samples are folded in half, wrapped in aluminium foil,
labelled and frozen for later analysis.
4, Before actual analysis the pads in aluminium foil are place'
in a drying oven overnight at 45°C.
5. Samples, standards and blanks are then loaded into sample
wheel and analysis begins.
Instrument; Control Equipment Corp. Model 240-XA Elemental Anlyzer
1. CHN Analysis - Carbon (CC^), hydrogen (I^O) and nitrog
) content in organic and inorganic compounds can be
ermined
(N-p)
dete
a. Combustion of the weighed or filtered sample occur
pure oxygen under static conditions (see figure ?)
Helium is used to carry the combustion products throug
analytical system to the atmosphere. Helium is also u:
for purging the instrument. It is a chemically inert <
relative to tube packing chemicals and has a high
coefficient of thermal conductivity.
-------
-------
-------
a. Solenoids A-G control the gas flow through the sys
valves H and I - are used for automatic leak test
The products of combustion are passed over suitable r<
in the combustion tube to assure complete oxidation.
reduction tube, oxides of nitrogen are converted to
molecular N and residual N2 is removed. Trie CC^/ wab
vapor and nitrogen are then flushed into a mixing volt
where they are thoroughly homogenized at a precise vo
temperature and pressure. This mixture is then releas<
the sample volume into the thermal conductivity detec
Between the first of three pairs of thermal conductiv
cells an absorption trap removes water from the saitipl
The differential signal read before and after the tra
reflects the amount of v/ater (hydrogen) in the origin
sample. A similar measure is made of the signal outj
second pair of thermal conductivity cells between whi
trap removes CO?. The remaining gas only consists of
nitrogen and helium. This gas passes through a therm
ccnductvity cell and the output signal is compared to
reference cell through which pure helium flows. This
the nitrogen concentration.
1. At the start cf each run, the entire system is flushe
helium at a high flow rate while the sample is in the
zone.
2. The injection box is automatically purged using the F
valve.
3. Tne combustion train is then filled with oxygen and t
sample is injected.
4. Shortly after sample injection, D valve closes to see
the combustion train from the rest of the analytical
system, which is still being flushed with helium.
5. Combustion occcurs under static conditions in an exce
oxygen at about 950°C.
6. During this tir.ie the mixing volume is being purged w:
and F valves open.
7. Then F closes to allow the pressure in the mixing vol
reach atmospheric pressure.
8. Close to the end of the combustion period, a high
temperature heat coil around the combustion tube vapc
any condensates at the entrance of the combustion tut
which may have been produced by diffusion of the samj
during initial stages of combustion.
-------
9. To assure complete combustion, the ladle is retracte
a small amount of 0 2 is added and the ladle is fully
injected.
10. During high heat, valve E closes, A and D reopen, an<
combustion products are completely flushed from the
combustion train into the mixing volume.
11. When a pressure of 1500 mm Hg is reached, valve D cL
trapping the sample gas in the mixing volume.
12. The time required to reach this pressure is called tl
fill time (usually 60-1GO seconds).
13. The combustion train remains under positive pressure
the end of the complete cycle.
14. While the sample gases are mixing, pure helium flows
valve C through the sample volume and through the
detectors.
15. The signal from each detector bridge is read and stoi
memory to provide a baseline reading with no sample c
the detector.
16. After mixing is complete and baseline reading has be<
F and G open which allows the sample gas captured in
mixing volume to expand through the sample volume to
atomsophere. During this time valve C is closed and
is low flow through the detector.
i. ~& i£. Ul££_S.uj;
17. Wnen sample gases are near atomospheric pressure, va]
and G close and C opens. The water, carbon dioxide c
nitrogen concentrations of the sample are Treasured b}
displacing the sample gas through the detectors to t\
atomosphere.
18. The volume of sample gas in the system is large enouc
that the helium flow allows measurement of the center
each detector in sequence, under steady state condit:
for at least 30 seconds.
19. The sample gas passses through the detectors at a cor
flow, pressure and temperature. This eliminates any
variation in water vapor pressure or water vapor
concentration due to changes in water adsorption of t
walls of the pneumatic system.
20. While the sample gas is displaced through the detectc
the output signals are recorded.
-------
21. T_h_e. dJJLf^jr-e-Rc_e. In microvcLts b_e_t_kLe_e_n .e_acJi
fur. iJie. aame. sL&ksfiiisj: JLs In
22.
23.
24.
At the end of a cycle, the exhaust valves are opened
allow the sample gases to escape to the atmosphere.
The HF-159 DATA HANDLER then prints out the calculate
results, places the instrument in STANDBY with C valv
open, and waits for the next command.
With the HA automatic injector the results are printe
after each run, but the run cycle continues until the
selected number of runs have been completed.
Definition £>JL
ELMKS
BOAT
CAPSULE
COMBUSTION
TIME
COMBUSTION
TUBE
DETECTOR
DETECTOR OVEN
DOUBLE DROP
FILL TIME
FURNACE
INJECTION
T_e_rjns_
Blank value = blank read minus blank zero.
An indicator of the stability of the system.
Platinum container used to inject sample into combi
furnace.
Aluminum, tin, or silver container. Used for seali
samples with an accurate weight, and maintains intec
prior to combustion.
Time for sample to fully combust in oxygen environn
Quart/ tube used for packing reagents and for sampj
combustion.
The heart of the analyzer consisting of triree brid
Determines the percentages of carbon, hydrogen, anc
nitrogen in the sample via thermal conductivity.
Keeps the temperature of the detector, pressure
transducer, mixing volume, and sample volume constc
On HA automation, two samples ore dropped for one i
used for filter and inorganic applications. Sample
requires a + prefix.
Time required to build up the pressure in the mixir
volume to 1500 mmHg.
Heats the reduction and combustion tubes to operat;
temperature.
Moving the ladle, containing a boat or capsule wit!
sample into the combustion furnace.
-------
INJECTION BOX
K-FACTOR
LADLE
MIXING VOLUME
MOTHER BOARD
READ SIGNAL
REDUCTION TUBE
RUN
RUN CYCLE
SAMPLE VOLUME
SCRUBBERS
TRAPS
ZERO VALUE
For the HA automation, the box assembly that houses
sample wheel.
Instrument sensitivity factor in microvolts per
microgram., calibrated using a chemical standard.
Transports the boat or capsule v/ith the sample into
combustion furnace.
Spherical bottle in which sample gases become homog
The main printed circuit board. All 240-XA power
supplies are located here.
Steady state signal produced by detector when sampl
gases are present in stable concentration.
Quartz tube with reduced copper that removes excess
from the sample gas and reduces oxides of nitrogen
free nitrogon.
One sample analysis from start to fir.ish, including
printout.
Typically a day of operation - the entire enalytica
sequence of runs frcm the first run to the last run
including the transfer of the run cycle data to the
Tube where sample gas is echausted from the mixing
prior to entering the detector.
Removes water and CO? from the gas supplies.
Used for removing water and C02 from the sample gas
Bridge signal with only pure helium flov/ing through
detector.
Calibration:
The following formula is used to calculate K factors, as well
and H concentrations in unknown samples.
%=!*!* (R-2-B) * 100
K W
where: K = Calibration factor for the instrument
W = Sample weight
P. = Read signal of sample gas
Z = Zero reading or instrument baseline
B = Blank signal (instrument, ladle and capsules)
U_s_e.d_: Acetanilide
-------
Composition: C = 71.09%
H = 6.71%
N = 10.36%
-tloner; The conditioner coats the walls of the system surfaces
(especially the mixing and sample volume) with water
vapor, carbon dioxide and nitrogen which simulate actu,
sample running conditions.
£la_nj<_s.: Should be run immediately after a conditioner.
i!vnf-a_c_Loj:_s_: Always run a conditioner before a standard and before a.
after a blank.
K factors vary greatly from instrument to instrument, b
should be within the following microvolt/microgram rang
KC = 15 to 25
KK = 44 to 76
K.N = 6 to 10
-------
Particulate Phosphorus (PP) :
The method used by C3L personnel is that of Aspila, et al. (1<
OUTLINE
1. Known volume of water passed through Whatman prcombus
mm GF/F filter (0.7 urn pore size).
2. Frozen
3. Dried at 50°C overnight
4. Muffled at 550°C for 1.5 hours.
5. Cooled overnight
6. Combusted filter placed in a labelled 50 ml plastic s
cap centrifuge tube and ] 0 ml IN H Cl added.
7. Capped and shaken several times during a 24 hour peri
8. Supernatent extract transferred to AutoAnalyzer cup w
Pasteur pipette.
9. Phosphate (that was extracted into the IN H Cl) analy
10. Blank filter pads are carried through the procedure a
Aspila/ I./ H Agemian, and A. S. Y. Chau. 1976.
serr.i-auton.ated method for the determination of
inorganic, organic and total phosphate in
t. 101:187-197.
Assembly: See figure ?.
Normal
Rat_e: 40/hour 9:1 Sample/Wash ratio
F_ili_e_j.: 880 nm
199-B021-04
Silicon at analysis temperature > 40°C and or
N H2S04 in the mixed reagent solution causes
interference in the concentration range of >
.05 mcj/ml silicon in the extract. These
conditions are avoided by maintaining an acid
concentration of 2.45 N ^304 in the reagents
analysis at 37°C.
-------
1. US
2.
Hydrochloric Acid (HC1) /
concentrated (sp. gr. 1.19) 86 ml
Deionized water (QS to :) 1000 ml
Add 86 ml cone. HC1 to approximately 800 ml good qua!
deionized water while cooling (cold water bath). Af
the solution is cooled, dilute to one liter with clei
water.
Add .5 g sodium lauryl sulfate (SLS) to 500 ml good
quality deionized water. Mix well!
3. SiilfiuJ^ Ac^Lo. JjL_i PJj.
Frcni orthcphosphate mc-thoc
4. Ajrjnojiiiin Mfilyt»jiAiL£j.
From orthophosphate method
5. LS££>JCbJ£. Ac.ig_i
From orthophosphate method
6. AxJJjRjmy P^-fcj^iS-uu:; T^jij:^i-e_
From orthophosphate method
7. SaiiiLUD Lauryl £ulLat-e. 1SL5L
From orthophosphate method
8. Ko_rJujici £e^aeJit^_L
From orthophosphate method
A. £i,ficJi 5iACldaxdj. From orthophosphate method
B. Secondary Standard; Take 0.1, 0.25, 0.5, 1.0 ml of £
standard A and dilute each to 100 ml with IN HC1 whic
yield 12 ug at P/l (.372 mg P/l); 30 ug at P/l (.93
P/l); 60 ug at P/l (1.86 mg P/l) and 120 ug at P/l C
P/l) .
-------
fin af.
[% on AA Chart of Blank] * F * HC1 extraction volume (ie.,
mg P/l = _________________________________________________________
Volume filtered (1) onto the filter pad
(F is the mean of ^
% on AA Chart of standard)
Total Suspended Solids (TSS) :
The method used by CBL personnel is basically that of AFHA met!
208D (Total Nonfiltrable Residue) dried at 1C3-I05°C and EPA method
Residue/ Total-Non-Filterable with some modification. Washing of f
pads with aliquots of deionized water has not teen included. TSS i:
retained material on a standard glass fiber filter disk after filtr.
of a well mixed sample of water. Results are expressed in rr.g/1.
1. V.'hatrr.an 47 mm GF/F filter pads (C.7 urn pore size) are
numbered and then weighed to 4 ce.cimal places.
2. The pads are then placed in an oven at 103°C for one ]
3. Padi- are then weighed.
4. In the field, a known volume of water is filtered thr<
the pad.
5. Upon returning to the laboratory, these pads are froz»
Day. af Analysis.
6. Filters are dried for one hour £t 1C3-105°C and then
weighed and the weights recorded. A few pads in that
are weighed again one hour later to check for any
additional weight loss. If there is more than a 0.5 r
weight loss between the same filter all pads are then
dried and re-weighed.
7. Calculation
(weight of filter + residue) - (weight of filter) * '.
mg TSS/1 = -----------------------------------------------------
mis of sample filtered
-------
0)
o
OL
CJ
i/l OJ
ro i—
3 U
ro
J- •*->
OJ D.
O. O
ro S-
1/1
E ^ ..
ro O CTi
<£
rx
a:
an
C
QJ
U
ro
Q.
to
V
a:
LU
CO
U
QJ
or
o E
_i c
o
o o
CO
CO
QJ
o
2
o
t/)
o
E -C c
E o. o
IT)
CD
I
QJ
X CM «£
o
E CQ
E I QJ
CT» 4~>
O cr> o
-------
APPENDIX D
METHODOLOGICAL COMPARISONS FOR NITROGEN DETERMINATION
IN ESTUARINE WATER SAMPLES
-------
Methodological Comparisons for Nitrogen and Chlorophyll
Determinations in Estuarine Water Samples
by
Christopher F. D'Elia1
Kenneth L. Webb
Diane V. Shaw1
Carolyn W. Keefe
^Nutrient Analytical Services Laboratory, Chesapeake Biological
Laboratory. Center for Environmental and Estuarine Studies, University
of Maryland, Solomons, MD 20688-0038.
*\
^Virginia Institute of Marine Science, Gloucester Point, VA, 23062
Submitted to:
Power Plant Siting Program
Department of Natural Resources
State of Maryland
Annapolis, Maryland
and
Chesapeake Bay Liaison Office
U.S. Environmental Protection Agency
Annapolis, Maryland
-------
ABSTRACT
This study was undertaken to compare results obtained with "standard" and
"alternative, new" techiques for total nitrogen and chlorophyll determination
in estuarine water samples.
The standard technique for total nitrogen (TN) determination recommended
by the U.S.E.P.A. involves the total Kjeldahl nitrogen (TKN) procedure in
which TKN + nitrate + nitrite gives TN. The EPA TKN procedure using the
Technicon Block Digester proved difficult to implement with estuarine water
samples: the block digestor heated samples unevenly and continous flow
analyzer baselines were unstable. However, standard "spikes" with a variety
of analytes yielded quantitative recovery and exhibited no salinity effect.
The alternative, the total persulfate nitrogen (TPN) technique, gives TN
directly and is easier to perform. More samples can be run per day using the
TPN procedure. TPN determination on standard spikes, like TKN, yielded
quantitative recovery and no salinity effect. A comparison of values obtained
using both techniques on natural, estuarine water samples collected from a
variety of locations in the Chesapeake Bay over an annual cycle yielded
equivocal results. The regression equation TPN (less nitrate 4 nitrite) =
21.79 (± 1.04) + TKN • 0.153 (± 0.021), best fitted the data. At low TKN and
TPN values the two techniques gave comparable results, but as TKN values
increased, TKN gave consistently higher values. Whether this discrepancy
results from an over-recovery by TKN or under-recovery by TPN cannot be
determined at present. Additional comparative work is continuing using a
modified TKN procedure to improve continous flow analyzer baseline stability.
The standard technique for chlorophyll a, determination recommended by the
U.S.E.P.A. involves grinding a glass-fiber filter, extraction with 90? acetone
and spectrophotometric determination of pigment concentration. The
alternative technique we tested involved extracting the filter with
dimethylsulfoxide(DMSO):acetone:water (9:9:2) and reading pigment
concentrations using a fluorometer calibrated with chlorophyll a. from a
commerical supplier. The results indicated that the fluorometric and
spectrophotometric methods for chlorophyll a. estimations in general use have a
low accuracy (approximately ± 30$) due to storage and interference problems.
The DMSO-based technique allows for the immediate extraction of pigments from
plankton samples and prevents the loss of chlorophyll a. due to storage and
subsequent grinding and extraction with 90$ acetone. In one comparison,
reduction in recovery after storage was nearly one-third. Chlorophyll b_,
which has been shown in the literature to interfere with the determination of
chlorophyll 3., was shown to occur in Chesapeake Bay phy top lank ton. For
convenience, cost, rapid extraction, and prevention of storage loss of
pigments, we recommend the DMSO-extraction technique followed by fluorometric
determination within several days. An acceptable alternative is to extract
and read the samples spectrophotometrically, within a few days of sampling in
cuvettes of appropriate path length (1-1Ocm), with and without acidification
for phaeophytin correction. If truly high accuracy, high precision results
are required, an HPLC method is desirable.
-------
Table of Contents
Overview 1-1
Section I - Comparison of TPN and TKN methods ....1-1
General Description of N Fractions in Natural
Waters ...1-1
Background and literature review 1-5
1. Wet oxidation procedures 1-5
a. Kjeldahl oxidation 1-5
b. Photo-oxidation ..1-5
c. Persulfate oxidation 1-6
2. Dry combustion procedures 1-6
Methods 1-7
1. Sampling and experiment s 1-7
2. TPN procedure 1-7
a. General description 1-7
b. Reagents 1-7
3. TKN procedure 1-8
a. General description 1-8
b. Reagents 1-8
c. Digestion procedure 1-8
d. Standards and blanks 1-9
4. Experimental comparisons 1-9
Results and Discussion .1-9
1. General observations 1-9
a. Block digestor temperature control 1-11
b. Standards 1-11
c. Teflon boiling chips '. ...1-11
d. Dilution loops 1-12
2. TPN and TKN recovery efficiencies vs
salinity 1-12
a. TPN 1-12
b. TKN 1-12
3. Comparison of TPN and TKN determin-
ations on estuarine water samples 1-17
4. Precision of TPN determinations on
replicate samples 1-17
5. Advantages and disadvantages of the two
methods. >...1-22
6. Further considerations 1-22
Summary and Recommendations 1-26
Section II - Comparison of chlorophyll methods ....II-l
General Description of chlorophyll rationale II-l
Background and literature review II-2
1. Calculations II-2
2. Interference by phaeo-pigments and
accessory chlorophylls 11-3
3. Storage, freezing »...II-5
Methods * .11-5
1. EPA Chesapeake Bay Study, July 1980 11-5
a. Sampling .11-5
b. DMSO extraction technique , ...II-5
c. Tube coating technique 11-6
-------
d. Fluorometry • II-6
e. Storage 11-6
£. Calculations 11-6
2. State of Maryland Chesapeake Bay
monitoring 11-6
3. Virginia EPA Chesapeake Bay monitoring 11-6
4. VIMS York River Plankton monitoring II-7
Results 11-8
1. Comparison of DMSO and 90% acetone for
extraction by fluorometry II-8
2. Comparison of Fluonnetry with spectrophotometry
a. 90% acetone with grinding.... 11-11
b. DMSO/Fluorometry compared with
90% acetone/spectrophotometry 11-11
3. Storage effects 11-15
4. Presence of chlorophyll _b and c^ 11-15
5. Precision of DMSO method... 11-16
Discussion 11-16
Comments on Interim Guidance on QA/QC for the
Estuarine Field and Laboratory Methods 11-19
Recommendations 11-20
References 11-21
Appendix I III-l
Appendix II 111-5
Appendix III III-8
Appendix IV ' III-9
-------
OVERVIEW
The following report is submitted jointly to the Maryland Department of
Natural Resources' Power Plant Siting Program (PPSP) and the Environmental
Protection Agency's Chesapeake Bay Liaison Office. The work reported on was
performed at the request of these agencies to compare (1) total Kjeldahl
nitrogen (TKN) determination using a semi-automated block digestor procedure
with a semi-automated alkaline persulfate nitrogen (TPN) digestion
determination and (2) several alternative methods of chlorophyll a
determination. These determinations are of considerable interest with
regard to water quality monitoring programs on the Chesapeake Bay. The TKN
vs. TPN comparisons were done in the Analytical Services laboratory of
Chesapeake Biological Laboratory (CBL) which typically uses the TPN
procedure, and the chlorophyll a^ determinations were performed primarily by
the Virginia Institute of Marine Science (VIMS) with assistance by CBL.
The funding agencies solicited this work to ensure that the adoption of
alternative, non-standard methods would provide data comparable to those
obtained using standard, EPA-approved methods.
SECTION 1
COMPARISON OF TPN AND TKN METHODS
General Description of N Fractions in Natural Waters
Figure 1-1 shows the nitrogenous fractions typically determined in water
quality studies. Also shown are the abbreviations typically used for these
fractions.
The distinction between "particulate" and "dissolved" nitrogen is
necessarily arbitrary. Particulate N (PN) is assumed to be that retained
on a filter having a nominal pore size between 0.45 and 1.2 um. Total
dissolved N (TON) is that passing through such filters, and undoubtedly
contains some small particulates and colloidal compounds, regardless of the
filter used. In most cases, the difference between that retained on
different filters in that range of nominal pore sizes is negligible,
although the filter matrix used may have an effect—organic "membrane"
filters are more prone to contamination than glass fiber filters.
.Figure 1-2 and Table 1-1 present all abbreviations used in this report
and give a comparison of how the different N fractions are determined using
standard EPA methods and the commonly used oceanographic measurements
employed by CBL. In Table 1-1 all determinations of a given fraction done
directly, i.e. not by difference or sum of other fractions, is indicated in
boldface.
The major differences between the standard EPA and commonly used
oceanographic procedures are that the latter (1) measure PN directly by
elemental (CRN) analysis of particulate material filtered onto glass fiber
filters, and (2) determine TON using alkaline persulfate oxidation (TPN
analysis). Oceanographers have adopted the alternate procedures for the
following reasons. Elemental analysis is extremely precise and offers the
1-1
-------
WHOLE WATER SAMPLE
TOTAL NITROGEN
(TN)
"ParticuI ate" Nitrogen
(PN)
Total "Dissolved" Nitrogen
(TDN)
"Dissolved" Inoranic Nitrogen
Nitrate
(N0")
Nitrite
(N02~)
nrnrron ' urn
(NH/)
Dissolved "Orcanic" Nitrcce
(DON)
FicLre 1-1. N fractions determined In typical water quality studies.
-------
A. Standard EPA
TN
0.45-utr. Killicore -e~;brane filter
PN TON
(TKN [whole water] - TKN [filtrate]) (TKN [filtrate] + N03~ + N02~)
DIN
DON
(N0
NH4+) ' TKN [filtrate] - NH4+)
N03 N02~ NK4
All by standard automated
colorimetrlc procedures
E. Typical Oceanographic (CBL)
TN
0.7-um GF/F class-fiber filter
PN
(Elemental Analysis
on f I Iter)
TON
(TPN [f iltrate])
DIN
NO,"
DON
(TPN [filtrate] - DIN)
NO;
NO-
All by standard auton-.cted
co1 orimetric procedures
fe 1-2. Ccrparison of stcridard EPA and typical ocearicgraphic
-------
Teble 1-1. Comparison of standard EPA and typical oceanographic (C3L)
procedures. Fractions measured directly ere boldfaced.
Fraction EPA Typical Oceanccraphlc (CBL)
TN TXN (whole water) . PN + TDN
+ N0~ + N0~
PN TKN (whole water) PN
minus TKN (f I Itrate)
TDN TKN (filtrate) + N03~ + N02" TPN (filtrate)
DIN K),~ + N09~ + NH/ • Sa^>e as EPA
H03~ (C-oIorlcetrlc) Ssrr« as EPA
Sarr>e as EPA
NH4+ (Colortnetrlc) Satne as EPA
DON TKN (filtrate) TDK minus DIN
minus (N03~ + N02~)
-------
advantage of .being a direct, rather than indirect determination of that
fraction. TPN digestion is much simpler and easier to perform than TKN
analysis, costs less to analyze per sample, and provides a direct
measurement of total dissolved nitrogen (TON).
Background and Literature Review
Oxidation procedures utilized in TKN and TPN methods are used
primarily to oxidize N-containing organic compounds, i.e. dissolved organic
nitrogen (DON). The following discussion pertains to these and similar
oxidation procedures for DON, and is provided here for general background
information. Much of this was exerpted from D'Elia (1983).
As was shown in Figure 1-2, DON is determined by difference between total
dissolved nitrogen (i.e. nitrate + nitrite + ammonia + organic nitrogen) and
dissolved inorganic nitrogen (i.e. nitrate + nitrite + ammonia) or by
aitterence between Kjeldahl nitrogen (ammonia + dissolved organic nitrogen)
and ammonia. A variety of oxidation procedures have been used to oxidize and
quantify DON.
I. Wet Oxidation Procedures
a. Kjeldahl Oxidation. Most of the earlier procedures for DON
determination lacked adequate sensitivity, and involved the traditional but
tedious Kjeldahl wet oxidation procedure (Kjeldahl, 1883). This approach
consists of an initial evaporation step followed by an oxidation with
concentrated sulphuric acid. It is generally regarded as difficult to
perform, and lends itself neither to shipboard use or to automation. In
early work, ammonium produced by the digestion process was determined by
titration (Barnes, 1959), while more recently colorimetric procedures have
been used (Strickland and Parsons, 1972; Webb et al. 1975; Webb, 1978). A
number of semiautomated procedures are in use in which samples are oxidized
by a manual Kjeldahl procedure with subsequent ammonia determination on the
digests being performed by autoanalysis using photometric (Faithfull, 1971;
Scheiner, 1976; Jirka et al., 1976; Conetta et al., 1976; Adamski, 1976) or
eiectrometric procedures (Stevens, 1976).
b. Photo-oxidation. The photochemical oxidation procedure first
developed by Armstrong et al. (1966) has generally superceded the Kjeldahl
oxidation procedure in most marine applications. A small quantity of
hydrogen peroxide is added to a sample contained in a quartz reaction
vessel, and high wattage mercury lamps are used to produce ultraviolet light
to photo-oxidize organic nitrogen, nitrite and ammonia to nitrate; nitrate
is then determined as described previously. The procedure is considerably
less tedious than the Kjeldahl procedure, can be performed at sea, and
unlike other procedures for DON oxidation, is relatively easy to automate
(Afghan et al., 1971; Lowry and Mancy, 1978). However, it does have some
shortcomings. Workers testing this method in freshwaters have found that
the photochemical reaction is very pH-sensitive and may not completely
oxidize compounds such as ammonia and urea (Afghan et al., 1971; Henriksen,
iy/U; Lowry and Mancy, 1978). Lowry and Mancy (1978) found that
ultraviolet digestion gave good results decomposing C-N but not N-N bonds,
yet felt that most compounds implicated in biological processes would be
recovered satisfactorily. Obviously, for samples containing a large amount
of nitrate plus nitrite, such as those from the deep ocean, the precision
ot DON determination by use of photo-oxidation will be less than that of a
-------
modern Kjeldahl procedure.
c. Persulfate Oxidation. Koroleff (1970; 1976) developed an
alternative wet oxidation procedure for total nitrogen determinations that
is becoming more widely used. He found that under alkaline conditions at
100°C and in the presence of excess potassium persulfate, organic nitrogen
in a seawater sample is oxidized to nitrate. Nitrate is then determined by
the standard photometric procedures used for nitrate determination. D'Elia
et al. (1977) and Smart et al. (1981) have shown that organic nitrogen
determinations by the persulfate and Kjeldahl techniques yield comparable
results and precision for both sea and freshwater samples; they also
discussed the advantages and disadvantages of persulfate oxidation relative
to Kjeldahl oxidation and photo-oxidation. Nydahl (1976) and Solorzano and
Sharp (1980) have suggested some improvements to Koroleffs original
procedure that alter reaction pH, lower blanks, and provide for the
requisite excess of peroxydisulfate. Nydahl (1976) noted that errors may
result when using persulfate oxidation on turbid samples; he also provided
an in-depth study of reaction kinetics and percentage recovery at varying
oxidation temperatures. Valderrama (1981) reported the simultaneous
determination of total N and total P using alkaline persulfate oxidation.
Goulden and Anthony (1978) have studied kinetics of the oxidation of organic
material using persulfate and have thus provided a basis for still further
refinement of the procedure such that simultaneous determination of C, N and
P may ultimately be possible on the same sample. As in the case of photo-
chemical oxidation, determination of DON by the persulfate technique will
have poor precision in the presence of large quantities of nitrate or
nitrite.
The original Koroleff procedure has been improved by Koroleff (see
Grasshoff et al., 1973) and modified recently to provide for increased
precision (Kalff and Bentzen, 1984) and for semiautomation and simultaneous
determination of both N and P (Gilbert et al., 1977; Ebina et al., 1983), and
tor determining N and P in particulate matter (Lagner and Hendrix, 1982).
Both reports indicated that satisfactory recoveries were obtained with most
organic nitrogen compounds.
l. Dry Combustion Procedures
Dry combustion procedures have been generally disappointing or
impractical for determining DON, although a recent report (Suzuki et al.,
iy85) suggests that a practical alternative may be at hand. Gordon and
Sutcliffe (1974) reported a dry combustion procedure in which a seawater
sample is freeze dried and the salt residues subsequently ignited in a CHN
analyzer. The obvious disadvantage of this is the need for a freeze drier
and the time involved in sample preparation. Other procedures have been
developed in which small volumes of sample are injected directly into a
combustion tube for evaporation and combustion (Van Hall et al., 1963;
Fabbro, et al., 1971; Hernandez, 1981), but these have not found wide use by
oceanographers because expensive and specialized equipment is required and
sea salt accumulation in the combustion chamber may reduce oxidation
efficiencies.
Recently, Suzuki et al. (1983) reported on a high-temperature
catalytic oxidation method in which nitrogenous compounds in liquid samples
are oxidized on a platinum catalyzer at 680°C under oxygen atmosphere and
the generated nitrogen dioxide (M) is absorbed into a chromogenic reagent,
-------
followed by a spectrophotometric determination. These authors report that
the TPN procedure yielded from 30-90% of the recovery afforded by their
pyrolysis technique. Unfortunately, the required instrumentation for this
procedure, the Sumitomo TN-200 total nitrogen analyzer is not available in
the U.S., and there have been no other published comparisons between results
of this dry combustion technique and wet oxidation procedures. However,
given the results of the Suzuki, et al. (1985) study, more comparisons
should be made between their dry combustion and other oxidation procedures.
Methods
1. Sampling and experiments. Samples for comparing TKN and TPN
determinations derived from three sources: (1) samples collected by the
"SONE" program of W.R. Boynton, et al.; (2) samples collected from the large
scale outdoor continuous culture system operated by the Academy of Natural
Sciences at Benedict, MD; (3) samples prepared in an experiment to compare
recovery of spikes of standard compounds in water of different salinity.
All samples were frozen as soon as possible after collection and
were thawed immediately before analysis.
2. TPN procedure. TPN determination was basically that of D'Elia et
al. (.iy//), with the following exceptions: (a) the oxidation was done on 10
ml samples in 30-ml glass screw-cap test tubes, and (b) the method used
to determine the nitrate concentration in the digest was the EPA-approved
AutoAnalyzer method (353.2)(USEPA, 1979).
This method with the above modification has been in use at CBL for the
past five years, although some improvements in the methodology have been
proposed by others (e.g. Valderrama, 1981; Sol6rzano and Sharp, 1980) that
may help further improve the method.
a. General Description. 15 ml of alkaline persulfate reagent is
added to the 10 ml sample in the 30-ml screw-cap test tube. Samples are
autoclaved at 100-110°C for one half hour and slowly brought back to room"
temperature. Each digested sample is neutralized by the addition of 1.5 ml
of 0.3 N HC1 and mixed with a vortex mixer. Two ml of borate buffer is then
added to the sample and vortexed. The nitrate concentration of the buffered
samples is then determined.
D. Reagents. Reagents were prepared as follows:
o Oxidizing reagent: 3.0 of NaOH and 6.7 g of low N «0.0003%)
potassium persulfate, K.,S20y, are dissolved in 1 liter with nitrogen-free
distilled water just before use.
o 0.3 N HC1
o Borate buffer solution: 30.9 g of HgBOo are dissolved in distilled
water, 101 ml of 1 N NaOH are added, and the solution brought to 1 liter with
distilled water.
-------
3. TKN procedure. We used a semiautomated total Kjeldahl nitrogen
(.TKN) procedure—EPA method 351.2 (colorimetric, semi-automated block
digestor, AutoAnalyzer II). The TKN procedure we employed was as close to
that used by the EPA's Central Regional Laboratory in Annapolis (U.S.E.P.A.,
iy/y) as possible. On several occasions, we used the identical equipment
used by EPA for analyses. This was done to obtain the most comparable TKN
data.
a. General Description. The sample is heated with a boiling chip
in the presence of sulfuric acid, potassium sulfate, and mercuric sulfate
for four and one-half hours. The residue is cooled, diluted to the original
volume and placed on the continuous flow analyzer for ammonia determination.
The determination of ammonia-N is based on a colorimetric method in which
an emerald-green color is formed by the reaction of ammonia with sodium
salicylate, sodium nitroprusside, and sodium hypochlorite in a buffered
alkaline medium at a pH of 12.8-13.0. The ammonia salicylate complex is
read at 660 nm using a continuous-flow analyzer photometer.
b. Reagents. Reagents were as follows:
o Digestion mixture: 25 ml Hg2S04 + 200 ml cone, sulfuric acid + 133 g
^SO^ are diluted to 1 liter with ammonia-free distilled water. l^SO^
solution: 8 g HgO + 10 ml cone. I^SO^ diluted to 100 ml with ammonia-free
DW.
o sui±uric acid solution (4%): add 40 ml of cone, sulfuric acid to 800
ml of ammonia-free distilled water, cool and dilute to 1 liter.
o Stock Sodium Hydroxide (20%): Dissolve 200 g of sodium hydroxide in
you ml of ammonia-free distilled water and dilute to 1 liter.
o Stock sodium potassium tartrate solution (20%): Dissolve 200 g
potassium tartrate in about 800 ml of ammonia-free distilled water and
dilute to 1 liter.
o Stock buffer solution: Dissolve 134.0 g of dibasic sodium
phosphate (Na2HPO^) in about 800 ml of ammonia free water. Add 20 g of
sodium hydroxide and dilute to 1 liter.
o working buffer solution: Combine the reagents in the stated order;
add 200 ml of stock buffer solution to 250 ml of stock sodium potassium
tartrate solution and mix. Add 120 ml sodium hydroxide solution and dilute
to 1 liter.
o Sodium salicylate/sodium nitroprusside solution: Dissolve 150 g of
sodium salicylate and 0.3 of sodium nitroprusside in about 600 ml of ammonia
free water and dilute to 1 liter.
o Sodium hypochlorite solution: Dilute 6.0 ml sodium hypochlorite
solution to 100 ml with ammonia-free distilled water (reagent is made
daily).
c. Digestion procedure. 20- or 25-ml samples are mixed well,
rinsed 3x with ammonia-free DW and the sample plus rinse water are added to
the digestion tube for each sample. 5 ml of digestion solution and 4-8
Teflon boiling stones are added to each tube, which is then mixed on a tube
-------
vortex mixer. With the block digestor in the "manual" mode/ the low and
high temperatures are set at 160°C and preheated until temperature is
reached (verified with a thermometer in sample of digestion solution alone).
Tubes are placed in digestor and heated at 160°C for 1 hour. After 1 hour
the "manual" mode is reset to 380°C and samples are heated for 2.5 hours
longer. At the end of 2.5 hours the block digestor is shut off manually.
Samples are cooled to room temperature at which time approximately 20-
ml of ammonia-free distilled water is added. Samples are then placed in a
sonicator (Astrason, Ultrasonic Cleaner, Model 13-H) for one-half hour to
break up precipitate. Each sample is mixed with a tube vortex mixer until
complete dissolution of all digestion residue and complete absence of layers
of solutions in the tubes. Ammonia-free distilled water is then used to
dilute samples back to the 25 ml initial sample volume.
During measurement of ammonia-N on the continuous-flow analyzer
(Scientific Instruments Corporation CFA 200) one set of reagents is used
during each sampling series. The continuous-flow analyzer is fitted with a
Kjeldahl manifold (Scientific Instruments Corporation TKN Cartridge No. 116-
540-0), which is used without the dilution loop (Figure 1-3). Reagent lines
are added to the manifold in the order: Working buffer, 4% sulfuric acid,
hypochlorite solution, and nitroprusside. The system is allowed to
equilibrate after the addition of each reagent and prior to running samples.
d. Standards and Blanks. TKN determinations included the following
standards and blanks:
o Ammonium sulfate standards: 0.0, 15.0, 45.0, 75.0 umol N L .
o Urea standards: 0.0, 10.7, 32.1, 42.8 umol N L"1.
4. Experimental Comparisons. We analyzed samples collected in the
tieJ-d and samples prepared in the laboratory to compare TPN and TKN recovery
efficiencies. Since TKN analysis yields organic nitrogen and ammonium
nitrogen and TPN analysis also determines nitrate, nitrite and ammonium,
direct comparisons cannot be made. Accordingly, we also performed nitrate
and nitrite determinations on all samples. The value obtained by
subtracting nitrate and nitrite from TPN is then comparable with TKN. Our
comparative studies included samples from: (1) The SONE program (August and
October, 1984; May, June, August, October, 1985); (2) An experiment in which
standards were added to samples of seawater diluted with distilled water to
different salinities; and (3) A wide range of N concentrations in the
outdoor large-scale continuous cultures at the Academy of Natural Science's
Benedict Estuar'ine Research Laboratory.
Results and Discussion
i. General Observations
TKN determination with the EPA-approved block digestor method proved to
be tedious and difficult. We chose to use this block digestion method because
it is often used when large numbers of samples must be processed and because
this is the method used by EPA in the monitoring program. We do not use this
procedure routinely in our laboratory, so much of our work was done at the
Central Regional EPA Laboratory in Annapolis, particularly until we were able
-------
u.
^
0
«
K
f
0
—
V-
O
>
oc
O
N
QC
•<
r«
n
o
' >
O
or
o
'•
it:
a
*
o
e
d
c
fl
a
0
• 2
ae
o
i
o
OB
O
Q
ae
O
w
oc
LJ
-J
Q.
Z
<
&
_l
>-
z
DC
O
(
' 5o
1 3-
£ 0<=
— P»
— UJ
Z -
VAJ
^—
<
>-
t.)
<
v>
C<
n
O
ac
— j
<->
0
a.
>-
•o
O
i
^
0
x
.^
a
_
Z
oc
o
UU
bo
<
£
O
~
i
>•
ce
O
5-
ee
O
t
i
Z
O
o
z
<
G
Figure 1 - 3.
fo' the sen iai
in Annapolis.
T.cfiifolcJ used in conjunciion wiin block digester
TKN procedure at E-A's Ceni^al cecicncl Lcboratory
-------
to gear up fully at CBL. We encountered a great number of problems
particularly with the digestion phase. The brand-new Technicon Block Digestor
we used failed to heat samples evenly and took a long time to reach
temperature. Analysts at EPA have also reported similar difficulties with
their block digestor. Once we had successfully determined block digester
preheating times and had calibrated the temperature regime achieved in each
individual position in the digestor, we encountered further problems. The
principal problem was with the use of the Teflon boiling chips recommended in
the EPA procedure. On samples containing appreciable salinity, at the latter
phases of the digestion procedure after most water had boiled off, the chips
floated and failed to prevent bumping and splattering. Such problems are
discussed in greater detail below.
a. Block digestor temperature control. Verification of exact
temperature settings and timing for the block digestor were made by filling
each heating cell with sand and measuring the temperature of the cells
during heating. The temperatures of selected cells were further verified by
measuring the temperature of a sample of digestion solution during heating.
Initially, the proper temperatures were attained and maintained by the
digestor according to the proper temperature schedule. However, when the
control was set on "automatic" the control box sporadically turned the block
heater off during heating, as well as boiled some samples dry (loss of
boiling chips and sample, which we termed "melt down"). Melt downs did not
appear predictable, i.e. they did not occur in the same block hole nor did
they occur during every digestion run. Samples were run on "manual" to
avoid the problems with the "automatic" setting. The occasional sample loss
due to melt downs could not be prevented. Due to these inconsistent
differences in temperature and melt downs between successive digestion runs,
standard curves based on ammonium sulfate and urea were constructed for each
set of samples digested.
b. Standards. The EPA Standard Operating Procedure for TKN
Determination recommends the following working standards of ammonium
sulfate: 0.2, 0.5, 1.0, 2.0, 3.0, 4.0, 5.0 mg N IT1. A standard curve of
these concentrations is non-linear at the higher concentrations and requires
a dilution loop. However, the concentration of total Kjeldahl nitrogen in
field samples is typically much lower than the lowest EPA standard (20 - 70
umol N/L) and the dilution loop, if used considerably reduces the analytical
precision of the TKN method. Due to the previous problems the following
standard curve was used: 0.0, 15.0, 45.0, 75.0 umol N L'1 (0.0, 0.21, 0.63,
and 1.05 mg N I/"*) based on an ammonium sulfate primary standard. Standard
curves were linear and field sample concentrations consistently fell within
this standard range.
The EPA procedure presents the data of one accuracy test which showed
100% recovery of organic-N from ammonium standards spiked with N-nicotinic
acid. Recovery of organic nitrogen depends upon the digestion history of
the sample, therefore each digestion run should include an accuracy test for
organic nitrogen recovery. For this reason each TKN run contained a urea
standard curve of 0.0, 10.7, 31.2, 42.8 umol N L'1 (0.0, 0.15, 0.45, 0.6 mg
N L-1).
c. Teflon boiling chips. The EPA method recommends cooling
samples 15 minutes, then adding water to the digestion tube up to the
initial volume before digestion (25 ml). The precision of estimation of
-------
ammonia-N is unavoidably affected because the boiling chips cannot be
removed from the samples before diluting to 25 ml.
d. Dilution loops. The standard Kjeldahl digestion manifold
(.Scientific Instruments, TKN Cart. 116-540-01) for ammonia-N determinations
dilutes each sample with distilled water in a dilution loop prior to the
introduction of reagents. Output curves recovered from the manifold with
the digestion loop appeared noisy with standards and samples almost
indistinguishable from background noise. Exclusion of the dilution loop
from the rest of the Kjeldahl manifold produced very distinct peaks for both
samples and standards (0.0 - 75.0 umol N L'1; 0.0 - 1.05 mg N L"1) which
were clearly above background noise. See Figure 1-3 for a diagram of
revised Kjeldahl manifold.
2. TPN and TKN Recovery Efficiencies vs Salinity
Once we had obtained satisfactory performance with our Kjeldahl
procedure, we performed the following experiment to compare TPN and TKN
recoveries at different salinities and concentrations. Low-nutrient,
continental shelf seawater and various dilutions thereof were spiked with
reference compounds (ammonium, urea, glutamic acid, and nitrate) at
concentrations ranging from 0 to 75 uM. The original data are presented in
Appendix II, with correlation coefficients for the standard curves in Appendix
ill. Precision of the total N determination by TKN and TPN taken from the
literature are compared by coefficients of variation in Appendix IV. For
future work with reference compounds, more-difficult-to-oxidize compounds such
as caffeine should also be tested (Suzuki et al., 1985).
a. TPN. Figure 1-4 shows peak heights obtained by the TPN (x-
axis) procedure plotted against seawater dilution (y-axis) and spike
concentration (z-axis). All peak height data are included for a given
percent seawater dilution and spike concentration, regardless of the
nitrogen compound used in the spike. Curves are fitted by eye to the
concentration data for a given seawater dilution—in effect, representing a
standard curve for each dilution. Precision is obviously good at all
seawater dilutions, and the "standard curves" appear linear.
Figure 1-5A through 1-5D present the percentage recoveries of spiked
compounds relative to nitrate standard curves in distilled water for the
same data lumped together in the previous figure. With the exception of
recoveries at the lowest spike concentrations which exceeded 100% (function
of ammonium contamination of the seawater used for the experiment that can
De corrected by subtracting a blank value determined for each salinity),
essentially 100% recovery occurred at all concentrations and dilutions.
To determine the upper range of the persulfate method, recoveries of
glutamic acid and urea were also determined on 150-750 umole spikes in the
given seawater dilutions. Essentially 100% recovery occurred at all
concentrations and dilutions.
b. TKN. Figure 1-6 shows peak heights obtained for TKN plotted as
a function of seawater dilution and spike concentration. As with the TPN
determination, there was no obvious salinity effect for the TKN procedure—
all standard curves clearly had similar slopes and intercepts on the y axis.
-------
CD
O
(O
c_
O>
u
z cr
Q_ o
03
a_
CLJ
Q_
C/3
Ndi
Figure 1-4. Three dimensional plot of TPN-detemined concentration of
standards (umol N L'l) vs. percent seawater vs. expected concentration of
spiked reference standards (umol N LT1).
1-13
-------
o
o
170*
Recovery vs Salinity
Ammonium by IPS
I
30
75
umoI/L Add«d
Figure l-5a. Percent recovery by TPN method vs. concentration of ammonium
(umol N L ) in different salinity water (a =• 0% seawater, b - 25 %
seawater, c » 50 % seawater, d = 75 % seawater, e » 100% seawater).
Kecovery vs Salinity
Nlh-aU by TPN
170*
Figure l-5b. Percent recovery by TPN method vs. concentration of nitrate
(umol N L"1) in different salinity water (a - 0% seawater, b = 25% seawater,
c = 50 % seawater, d = 75 % seawater, e • 100 % seawater).
1-14
-------
o
o
o
o
17OS
Recovery vs Salinity
Glutamlc Acid by TPN
25.2
75.5
umol/L Add«d
Figure l-5c. Percent recovery by TPN method vs. concentration of glutamic
acid (umol N L,"1) in different salinity water (a - 0% seawater, b - 25%
seawater, c » 50 % seawater, d - 75% seawater, e = 100% seawater).
Recovery vs Salinity
Ur»a by T7»K
170S
76.8
53.6
umol/L Add«
-------
o
o
ro
c.
OJ
u
a- T3
O C
ro
j_i
o -u
r-t C
a. QJ
u
c_
QJ
a_
in
Figure 1-6. Three dimensional plot of TKN-determined concentration of
standards (umol N L~l) vs. percent seawater vs. expected concentration of
spiked reference standards (umol N L ).
-------
However, precision clearly was not as good by TKN as it was for TPN, and as
expected for the procedure, nitrate was not recovered. The nitrate points
are connected by additional lines fitted to the data.
Figures 1-7A through 1-7D presents the percentage recoveries of the
individual spiked compounds relative to ammonium standard curves in
distilled water analyzed by the TKN method. Clearly the precision was less
than for the TPN analysis, but recoveries appeared complete at all
salinities and spike concentrations. However, a small amount of nitrate
appeared to have been recovered in some samples—this is anomalous
because TKN should not reduce nitrate to ammonium, and is probably
explained by contamination. Nonetheless, there is the interesting prospect
of some unexplained nitrate reduction occurring, which would be difficult
to explain chemically.
.}. Comparison of TPN and TKN Determinations on Estuarine Water Samples
Samples over a range of salinities were collected from August, 1984
through December, 1985 for comparison of results obtained using TPN and TKN
determinations. These data were obtained from the "SONE" monitoring program
conducted for the State of Maryland and in large-scale continuous cultures
drawing water from the mesohaline region of the Patuxent River.
The results of these comparisons were poor and the explanations for the
j.acK, ot comparability between TKN and TPN - nitrate + nitrite (comparable
values) is as yet unresolved, despite exhaustive checking and rechecking of
dii procedures and calculations. We wish it were as simple as having
ignored that ammonium sulfate standard has two moles of N per formula
weight, but we did not make that error . We also are aware that refractive
index problems can affect results (Froelich and Pilson, 1978) and that pH
adjustment of the acid digest is critical for proper color development
(.Keay, 1985). Figure l-8a shows the comparison of data from digestions we
deemed "good" according to the criterion of low rates of bumping and
splattering. Figure l-8b shows the comparison of data from all digestions
and determinations we performed. While comparisons of samples containing
less than 30 uM Kjeldahl nitrogen seem close, there appears to be a
systematic difference between the two procedures. The regression equation
best fitting this relationship is: TPN - N023 = 21.79(+1.04) + TKN*0.153
V+U.021). it is not clear from this study whether the discrepancy between the
iPN and TKN data in Figs. 1-8 and l-8b is "real" or due to a contamination
problem.
4. Precision of TPN Determinations on Replicate Samples
The CBL nutrient analytical services laboratory has been conducting TPN
analyses for the bay-wide EPA-sponsored monitoring program since May, 1985.
These analyses are conducted over a wide range of salinities and total
dissolved nitrogen concentrations and are subjected to a rigorous QA/QC
protocol, as dictated by EPA. To illustrate the achievable precision of the
TPN determination on duplicate samples (each involving separate filtration,
aliquoting and storage), it seemed appropriate to present here the results
from the QA/QC program. Figures 1-9A and 1-9B show the EPA QA/QC plots for
standard deviation of duplicates vs. mean concentration and for coefficient
of variation vs. mean concentration. The mean coefficient of variation for
all samples is approximately 8%, an excellent value considered that it
represents more than analytical error alone. Typical coefficients of
-------
22OX
Recovery vs Salinity
Ammonium by TKN
o
o
13
O
200?: -
1BOX -
16OX -
14035 -
120?:
1OOX -
30
75
umol/L Add«d
Figure l-7a. Percent recovery by TKN method vs. concentration of ammonium
(.umol N L ) in different salinity water (a = 0% seawater, b = 25% seawater,
c - 50% seawater, d =• 75% seawater, e - 100% seawater).
Kecovery vs Salinity
Nttrctrs by TKM
170X
o
o
e
>
150X -
140X -
130X -
120X -
11 OX -
100X -
SOX -
SOX -
70X -
60X -
SOX -
•cox -
SOX -
20X -
10X -
ox
—r~
23
1—
50
umol/L ^
—I—
75
Figure l-7b. Percent recovery by TKN method vs. concentration of nitrate
(umol N L in different salinity water (a - 0% seawater, b = 25% seawater,
c - 50% seawater, d = 75% seawater, e = 100% seawater).
1-18
-------
170*
Recovery vs Salinity
Clutamlc Acid by TKN
o
o
3.
B
O
25.2
50.4.
75.5
umol/L Added
Figure l-7c. Percent recovery by TKN method vs. concentration of glutamic
acid (.umol N L ) in different salinity water (a • 0% seawater, b = 25%
seawater, c » 50% seawater, d - 75% seawater, e = 100% seawater).
Kecovery vs Salinity
Urea by TX>4
170*
o
u
3.
o
>
80.4.
Figure l-7d. Percent recovery by TKN method vs. concentration of urea (umol
N L"1) in different salinity water (a » 0% seawater, b = 25% seawater, c -
50% seawater, d - 75% seawater, e - 100% seawater).
1-19
-------
r
r
I
r
o
100
90 -
SO -
70
60
50
XO
30
20
10
TKN vs. TDN-(Nitrate
"Good" Xna(y»«»
Nitrite)
20
60
80
100
TKN
Figure l-8a. TDN - (nitrate -t- nitrite) vs. TKN determinations of estuarine
samples for analyses without bumping and splattering ("good" data).
TKN vs. TDN-(Nitrate + Nitrite)
All Data
E
Z
z
a
100
BO
70
60
50
30
20
10 -
20
I
EO
I
SO
100
TXN
Figure l-8b. TUN - (nitrate + nitrite) vs. TKN determinations of estuarine
samples for all analyses preformed.
1-20
-------
0.24
TPN Field Duplicates
May, 198.5-Jen, 1988
0.22 -
0.2 -
0.1 B-
0.16-
0.14-
I °'12H
• 0.1 -
o.oa -
0.06-
0.04-
0.02 -
v
4 ;
11 1 i I I i I
O.3 0.5 0.7 0.9 1.1
U«cn of
i i i r r
1.3 1.5
i r i i
1.7 1.9 2.1
figure l-9a. Standard deviation of duplicates vs. mean concentration of
field duplicates for bay-wide EPA-sponsored monitoring program.
40
TPN Field Duplicates
Wcy. 1985— Jan, 1B88
c
o
o
o
"o
I
o
o
o
33 -
30-
23 -
20 -
13 -
10 -
3 -
r — r" i - 1
T r—i r
O.3 0.3 0.7 0.8 1.1 1.3 1.3 1.7 1.9 2.1
Figure l-9b. Coefficient of variation vs. mean concentration of field
duplicates for bay-wide EPA-sponsored monitoring program.
1-21
-------
variation for Kjeldahl analyses are given in Appendix IV.
5. Advantages and disadvantages of the two methods.
while this work has clearly not shown the equivalence of the two
analytical determinations, we believe that our analytical inexperience with
the TKN procedure and the poor semiautomated TKN protocol are responsible
for the lack of comparability. We recommend that further comparisons be
made between TKN and TPN determinations. In addition, we also recommend
that a laboratory that routinely runs TKN analysis, not with the block
digestor, split samples with us, so that we can do TPN determinations for
comparison.
It is important to emphasize why it is worthwhile to pursue the
comparative work further. TPN analysis offers a number of advantages over
Kjeldahl analysis that make it a highly desirable alternative to TKN. Such
advantages in cost, ease of use, and excellent precision (cf. Fig 1-9A and
1-9B) means that TPN determination deserves further comparison.
Table l-II shows the analyst's time and steps involved in processing a
series of TKN samples. Table l-III shows a comparison of the analyst's time
and steps involved in processing a series of TPN and TKN samples.
Table l-III summarizes the advantages and disadvantages of the two
procedures.
6. Further Considerations
Although there have been reports by Japanese workers that the alkaline
persulfate digestion technique substantially underestimates total nitrogen
in seawater compared to the oxidative pyrolysis technique, several points
should be made regarding comparability between the two methods. First,
results have not been reproduced by others, probably due to the
unavailability of the Japanese instrument in other countries. Secondly,
while the Japanese workers did not state the temperatures at which their
oxidation was carried out, the temperature used may have exceeded that
recommended for optimum digestion. Goulden and Anthony (1978) and others
have cautioned that high temperatures will cause too rapid a breakdown in
the persulfate and poor oxidations.
One criterion that Suzuki et al. (1985) used in criticism of the
persulfate technique was that it yielded poor recoveries of caffeine.
However, B. Nowicky and M. Pilson (pers. comm.—cf. Appendix I) have
obtained complete recovery of nitrogen in caffeine.
The persulfate oxidation procedure could be optimized still further—
especially worth checking are (1) the heat of combustion and speed with
which the samples are brought up to temperature, and (2) the ability of the
procedure to oxidize complex rings.
-------
Table l-II. Comparison of analyst's time and steps required for the
and TKi^ methods.
Method Day Step and Activity Time Involved
(hours)
TPtt 1 1. Thaw 10U samples (1U ml in 30-rnl tubes)
2. Make up standards and put in 30-ml tubes. 0.4
3. Make up 2 L"oxidizing reagents. 0.1
4. Add 15 ml oxidizing reagents to all
standards and samples. 1.0
5. Autoclave at 100 - 110 degrees C. . 0.5
6. Cool in autoclave. 1.0
7. Kemove fron autoclave and cool to room
temperature. 1.0
8. Make up 0.3 N HC1 and borate buffer. 0.1
9. Add 1.5 ml 0.3 N HC1 and vortex mix. 1.0
10. Add 2.0 ml borate buffer and vortex mix. 1.0
2 1. Set up continuous flow analyzer. 1.0
• 2. Prepare and run nitrate standard curves. 0.5
3. Run samples and standards. 3.0
' 4. Shut down auto analyzer. 0.5
5. Read charts and calculate concentrations. 2.0
6. Wash tubes and caps. 1.5
Total 14.6
"Time/Sample 9 min
1-23
-------
Table l-II, cont'd.
Met nod Day Step and Activity Time Involved
(hours)
TKN 1 i. Thaw 45 samples (2U-25 ml in 30-nil tubes)
and put in Kjeldalil digestion tubes. 0.4
2. Prepare ammonium standards. U.I
3. Put 25 ml samples and standards in
Kjeldahl digestion tubes. 1.0
4. Add 5 ml digestion solution to all
standards and samples. 0.25
•5. Add 2 boiling chips to each sample and
vortex mix. 0.25
6. Digest standards and samples in clock
digestor at the following temperatures
and times:
Temperature (degrees C)
90 0.25
120 0.5
150 0.5
180 0.5
200 0.5
230 0.5
360 2.5
7. Let cool in digestor. .1.0
8. Remove from digestor and cool to room
temperature. 2.0
9. Dilute cooled samples and standards to
25 ml with distilled water and
vortex mix. 1.0
1 or 2 10. If solid develops and persists'after
dilution to volurae, sonicate covered
samples to break up solid, then allow
samples to settle. 2.0 to 3.0
2 1. Set up continuous flow analyzer. 1.0
2. Run digested ammonium standard curve. 0.5
3. Run digested samples in duplicate. 2.0
4. Shut down continuous flow analyzer 0.5
5. Read charts and calculate concentrations. 2.0
6. Wash tubes and caps. 1.5
Total 20.75
Time/Sample 28 rain
1-24
-------
Table 1-1 I I. Comparison of the TKN and TPN methods for the procedures we
used and assuming the availability of an autoanalyzer colorimeter, sampler,
pump and chart recorder.
Characteristic or Feature
TKN
TPN
Estimated Cost
Startup
Block Dlgestor
Pressure Cooker
Autoanalyzer manifold
$504
$3395
$1000
$250
$ 80
$430
Total
Per Sample Charge In our
Laboratory
Special Equipment
Ease of Use
Samples per Day
Precision (CV$)
$18.00
Fume Hood
Block Dlgestor
AutoAnaIyzer
Kjeldahl Tubes
Not easy
20
$5.75
Pressure Cooker
AutoAnaIyzer
. Test tubes
Very Easy
50
~3*
Comments
Seawater samples
are more difficult
— proper boiling
chips must be used
DON not precisely
determined In the
presence of high
nltrete concentrations
-------
Summary and Recommendations;
1. The persulfate total nitrogen procedure is easier to perform, yields
better routine precision, requires less expensive and sophisticated
digestion apparatus, and requires less analyst time per sample. This
procedure deserves further evaluation as a potential standard digestion
procedure for total dissolved nitrogen by EPA.
2. Both methods yielded expected and complete recoveries of laboratory-
spiked samples over a wide salinity range. However, results obtained
comparing natural estuarine samples appeared to yield a systematic
difference between the two procedures that is as yet unresolved.
J. The block digestor for the TKN procedure does not perform well and proved
difficult to use, particularly in the hands of technicians inexperienced in
its use. Differential heating of different locations on the digestor must
be accounted for. The heating characteristics of the digestor seem to
depend on external factors such as location in the hood, laboratory
temperature and warm-up time. Such factors need to be accounted for if the
block digestor is to be used.
4. The residue remaining in the digestion tubes after block digestion of TKN
samples is very difficult to redissolve in high salinity samples. Sonication
may be required as well as long sitting times. Contamination may occur during
such sitting times. A better re-dissolution procedure should be developed for
high salinity samples.
3. Additional comparisons should be made between the two procedures using
split samples from the natural environment. We recommend that a laboratory
not using the block digestor and achieving TKN results satisfactory to EPA
share samples with us so that we can perform additional TPN analyses.
b. Organic N standards in seawater should be used for standard curves.
Such standards should include difficult-to-oxidize nitrogen-containing
reference materials, e.g. nicotinic acid, caffeine.
-------
SECTION II
COMPARISON OF CHLOROPHYLL METHODS
General Description of Chlorophyll Rationale
Many aquatic investigations utilize one or more estimates of
photoautotrophic plankton biomass, e.g. cell counts, total cell volume
estimates, protein determinations, dry weight, cell carbon, nitrogen,
phosphorus or silica and pigment analyses including chlorophyll a_
determinations. The use of chlorophyll &_, especially fluorometric
determinations, has become widespread, possibly to the point of
indiscriminate use, because the method is relatively fast, simple and
reproducible. The use of this biomass measure has been questioned
because it may vary by an order of magnitude relative to other biomass
measures, e.g. dry weight, cell volume or cell protein. Eppley (1977)
reported 10-fold variation in cell carbon:chlorophyll a_ ratio of
phytoplankton. The failure of the fluorometric method to provide any
information about population structure as well as the observed
interference problems from accessory pigments and phaeo-pigments are
largely overlooked.
Any monitoring or other routine sampling program for chlorophyll
pigment must address certain criteria such as: (1) design of sampling
scheme, e.g. frequency, depths, replicates, etc., (2) technique of
sampling, e.g. by pump, bottle, rossette sampler, etc., (3) sample
treatment, e.g. filtration, including types of filters and filter holders
or the use of whole unfiltered water samples, (4) possible storage of
samples either before and/or after filtration or extraction,, (5)
extraction techniques including solvent composition, temperature and or
physical treatment (sonication or grinding) and duration of extraction,
(6) quantification method such as spectrophotometric, fluorometric or
spectrofluorometric determinations on the gross extract, and (7) how the
calculations are made after the raw data are gathered.
Recently a variety of solvent systems containing dimethyl sulfoxide
(DMSO) has been suggested for the extraction of chlorophyll type pigments
from freshwater phytoplankton (Shoaf and Lium, 1976; Stauffer et al.,
1979). Burnison (1980) has described a method using pure DMSO at 65 C
followed by dilution with 90% acetone; Speziale et al. (1984)
subsequently compared this method to N,N-Dimethylformamide (DMF) and 90%
acetone extractions on natural samples and cultured freshwater
phytoplankton. Both DMF and DMSO were better extractants than 90%
acetone, with DMF being very slightly better with chlorococcalean
species. No work has been published concerning the use of DMSO:acetone
solvent systems with marine plankton species, although Seely et al.
(1972) reported using DMSO as part of a serial extraction method for
brown algae and a modified method is suggested for marine macrophytes
generally (Duncan and Harrison, 1982). Although there is reason to
predict that DMSO:acetone solvents are more effective in extracting
marine samples than present acetone methods, the method should be
evaluated before it is utilized extensively. We have recommended a DMSO
technique as the procedure of choice for the EPA-Chesapeake Bay
Monitoring program because it is easy, requires a minimum of handling,
II-1
-------
storage as a separate step isn't required, and it gives results identical
to the 90% acetone extraction with grinding for an uncorrected (for
phaeo-pigments) chlorophyll £ value by fluorometry.
The original scope of this work was to further investigate
extraction techniques for chlorophyll _a; it was expanded to include some
aspects of sample storage (freezing) and a comparison of
spectrophotometric and fluorometric determinations in order to assist the
interpretation of the data.
Background and Literature Review
1. Calculations:
Methods manuals (e.g. APHA, 1985; ASTM, 1979; Parsons et al., 1984)
appear to be in consensus that the accepted methods for spectrophoto-
metric determination of chlorophylls involves the use of the trichromatic
equations of Jeffrey and Humphrey (1975). The spectrophotometric
determination of phaeo-pigments utilizes readings taken at 665 or 664 nm
before and after acidification and the formulae of Lorenzen (1972) for
the calculations. The formulae for a 1 cm cell are as follows:
Jeffrey and Humphrey (jug chl/ml extract for 1 cm cell)
Chlorophyll £ = 11.85 E(at 664nm)-l .54E(at647nm)-0.08E(at 630nm)
Chlorophyll _b = 21.03 E(at 647nm)-5.43E(at664nm)-2.66E(at 630nm)
Chlorophyll c_= 24.52 E(at 630nm)-1.67E(at664nm)-0.08E(at 760nm)
where E is the absorbance at different wavelengths corrected by a blank
reading at 750 nm. Chi per unit seawater is then calculated by:
Chlorophyllyug/l = (Chi x v)/V
where v is the extract volume in ml and V is the sample volume in liters
Lorenzen (for 1 cm cell)
Chlorophyll a. (jug/1) = [26.7(665b-665a)v]/V
Phaeo-pigments (jug/1) = [26.7(1.7(665a)-665b)v]V
where 665a and b are after and before acidification respectively and V
and v are as above. The b reading is listed at 664 in APHA (1985) and
ASTM (1979), while the original articles (Lorenzen, 1967) and Parsons
et al., 1984) cite 665nm for both the b and a readings. In this
presentation we use the above equations although Speziale et al. (1984)
indicates that the Lorenzen equations cause underestimations by about 6%,
i.e. the 26.7 of the above equations should be replaced by 28.4.
The above equations are often utilized directly from manuals
without consulting the original volumes. Thus, one may not realize
that Jeffrey and Humphrey published four sets of equations, for differing
kinds of populations: 1) Chi a^ and ^ for higher plants and chlorophyta,
2) Chi ji and cl, c2 for diatoms, chrysomonads and brown algae, 3) Chi a_
and c2 for dinoflagellates and cryptomonads, and 4) the above equations
for mixed populations of phytoplankton. Chi ji was well recovered by all
equations (98-102%). The specific equations for a + b and a + c gave
similarly good values for all the pigments, however the mixed plankton
equation gave good results for b and c only when these pigments were
abundant relative to chl £, i.e., a:b or a:c ratios of less than 4:1.
II-2
-------
2. Interference by phaeo-pigments and accessory chlorophylls:
The use of all of these equations assumes that the solution
analyzed is a mixture of pure pigments and contains no decomposition
products. The colored phaeo-pigments, Table II-l, in contrast to the
colorless ones, show up in these data as chlorophyll a. Prior to 1978,
Table II-l. Chlorophyll breakdown products (phaeo-pigments)
Phaeophytin a
Chlorophyllide a
Phaeophorbide a
Absorption
peak
667nm
664nm
667nm
Absorption
coefficient
51.2
127
74.2
Reference
Score
Score
Score
phaeophytin a^ was thought to be found only in traces in natural marine
samples; this was subsequently found not to be true. Pheaophytin
-------
01 01
LJ -^
-0 »
a -M
&IA
—
*; ai ^.
a ai
S--S "£
a •£> -a
" in u
> « r
a g ->-•
.£ «•• •
_ a
0 *
a a.
j= "5 a
g- S J
tj £ *
*!*
-0^0
** • **
U ^ B
•£ ^ S
£ " I
J T g
hJ>. *^
•S g2
S**4 * *
—«
n w ai
S" -3 E
" fe fc «
*J >*. -M
. —• k. B
CM —« 01 0»
I -u •
« js -5 •«
-»j a.
ai in o
3 a o •«
•o k. j= js
i— •*• at a.
>*• c=
g
I
Sirt in
CM .
ra . O
I I I I I I
OP«.J»—<[gro«p».
ip -o o CM r> _.
— . — « CM
—*r—.•^•.OP^^C/.^CM^O*
*»—^*^*^—^—*o^o^^
O'O — <=>O<=0>OG>
II 1 III
I I I I
' CM "^ in ** °*
CM -O
I I I I I I
I 1 I I I I
CM
^ ro irj ^
i • - • • i
2 R - -
1 1 1 1
i i i i i i i
i S
r~ K)
in
o
aa BO
-^CNiroirt-or-^aoff-^-^txro^irt
o* ai
**•• A
s -a
SOI ^j M
mi 0; k. u *.
.„ -O k. 01 C
^j Ol •*« O k. Ol
<• -o — y k. •
.ii 2 *• § S -2"
^ 5 "I- «• ••' g
S O' o !!B !S J=
-a at —« cj «j a.
"" a. u u ii u
-------
3. Storage, Freezing:
The effect of storage conditions on chlorophyll determinations are
not well documented in the literature. Most methods use magnesium
carbonate on the filters to prevent acid conditions from causing
chlorophyll degradation. The recommended DMSO method uses 0.1% by volume
of diethylamine to maintain alkaline conditions. Jeffrey and Hallegraeff
(1980) froze filters in liquid nitrogen and then held them at -20C until
extraction. This method resulted in a 5-10% loss of chlorophyll a_ in 6
weeks of storage with a gain of 2-3% phaeophytin, presumably the major
breakdown product was colorless.
Some publications suggest that stored extracts or extracting tissue
show less degradation of chlorophyll than do plankton samples stored on
frozen filters. For example, Wood (1985) reported 11-21% loss of
chlorophyll from samples stored dry when compared to those stored in
extracting solvent for 9 days. Similarly Moran and Porath (1980),
reported no loss of chlorophyll in N,N-Dimethylfonnamide with dark
storage at 4C. Inskeep and Bloom (1985), however, reported no difference
between stored soybean leaf disks with and without solvent. Logic
suggests that extracting solvents such as DMSO may denature enzymes
which denature chlorophyll and that, consequently, combinations of tissue
and extracting solvent may remain stable for chlorophyll concentration
even at room temperature.
Methods:
1. EPA Chesapeake Bay Study, July 1980
a.) Sampling. Samples for the extraction method comparison, between
DMSO and 90% acetone with grinding, were taken from a field study in the
York River (USA) (37'15'40" N. Lat, 76'23'28" W. Long) and from 4
stations on a transect across Chesapeake Bay along Long 37' 20', July 8-
16, 1980. These field samples consisted of the surface samples (1m
depth) processed by standard fluorescence methods (Yentsch and Menzel,
1963) with freezing for less than a week, in triplicate (and were a
subset of a larger sample set) and additional samples in duplicate from
the 1 m water samples for extraction with dimethylsulfoxide (DMSO):
acetonetwater (9:9:2) with 0.1% by volume of diethylamine (DEA); insofar
as possible the samples were taken twice a day at the five stations for
9 consecutive days. Whatman GF/F filters were used because they retain
more chlorophyll than a number of other filters tested.
b.) DMSO extraction technique. A measured volume of sample
sufficient to produce visible color on the filter disc was filtered
through a Whatman GF/F 2.5 cm filter. For estuarine water 5-10 ml is
usually sufficient. The filter was folded with the sample side inward
and placed in a 16x100 mm glass culture tube which had been coated (see
below) to exclude as much light as possible. The tube contained a 10 ml
aliquot of DMSO and a minimum of air space. The tube was closed with a
teflon lined screw cap and the filter was extracted for at least 2 hours
at ambient temperature. Filters were always manipulated with forceps.
It was not necessary to filter or centrifuge the sample before measuring
fluorescence.
II-5
-------
c.) Tube coating technique. To exclude light from the culture tubes
during extraction, the tubes were dipped twice in a mixture of lampblack
and plastic "tool grip compound" obtained from Brooks tone Company,
Peterborough, NH. About 70 cc of lampblack was added to each 16 oz. can
of red compound and mixed thoroughly. Approximately three dozen tubes
were coated from each can.
d.) Fluorometry. Fluorescence measurements were taken with G.K.
Turner Associates Model 111. Purified chlorophyll a_, (Sigma Chemical
Company, product no. C-5753, lot number 39C-9690) was used for
calibration. Concentrations were verified spectrophotometrically using
the equations of Jeffrey and Humphrey (1975). Spectrophotometric
measurements were taken with a Bausch and Lomb Spectronic 710. The Sigma
standard was dissolved in 100% acetone and then diluted so that final
concentrations of solvents matched those of the extraction systems.
e.) Storage. To test the effect of storage on extracted material,
a second repetition of some of the DMSO samples were extracted in the
original sample tubes at room temperature for varying periods up to 32
days after the first repetition was read.
f.) Calculations. The pigment concentration (jig 1-1) values were
calculated as follows: (1) uncorrected (for phaeophytin a) chl a_
equivalents directly from before acidification fluorescence values
(Strickland and Parsons, 1972, page 201) and (2) corrected chl a_ and
phaeophytin from the before and after acidification values (Yentsch and
Menzel, 1963). Because sample variance was significantly correlated with
sample mean, a log transform was performed before analysis (Snedecor and
Cochran, 1967, page 329). All statistical analyses were performed using
Statistical Analysis System GLM, CORR, SUMMARY,and MEANS procedures (SAS,
1979).
The comparisons were made on paired sets (i.e. data from two methods
on the same water sample) in duplicate, the duplicate values for the
standard method'were produced arbitrarily by choosing the first two
values in the data set from the existing triplicate values. The second
of the DMSO duplicates was analyzed in a time series fashion, i. e. 0, 1,
2, 10, 16 or 32 days after its pair, in order to allow testing for
extraction time/storage time effects.
2. State of Maryland Chesapeake Bay Monitoring
Approximately 80 samples were collected for chlorophyll analysis on
each of five cruises (August and October 1984 and May, June and August
1985) for a total 388 individual samples. At each station samples were
taken from two depths, surface and bottom, in quadruplicate. Sample
volume varied from 50 to 1000 ml depending upon the apparent chlorophyll
in the sample. Samples were filtered onto 47 mm Whatman GF/F filters
and frozen for the duration of each cruise, 1-5 days. Two of each set
of replicates were analyzed by the CBL laboratory following the DMSO
extraction technique described above but starting with frozen samples.
The two remaining replicates from each station were kept frozen and
transported to the Virginia Institute of Marine Science (VIMS) for
analysis by the method (Strickland and Parsons, 1968) of grinding in 90%
II-6
-------
acetone, allowing to stand overnight in the refrigerator, centrifuging
and reading on either a Turner Model ill or Turner Designs fluorometer.
Most extracts were sufficiently concentrated to be analyzed by
spectrophotometry; such was done using a 1-cm cell in a Gary Model 15
spectrophotometer. Spectrophotometric readings were taken at 750, 665,
664, 647, 630 nm and at 665 nm after acidification. The trichromatic
equations of Jeffrey and Humphrey (1975) were used to calculate
chlorophylls a^ _b, and c_. The assumption is made that no phaeo-pigments
are present when these equations are used. Chlorophyll a_ and
phaeo-pigments were also calculated with the 750 nm and the 665 nm before
and after acidification readings by the equations of Lorenzen (1967).
Chlorophyll b^ interferes with this evaluation.
3. Virginia EPA Chesapeake Bay Monitoring
We accompanied the VIMS Bay monitoring cruises on 8 consecutive
cruises from mid-April through mid-August 1985. Sampling procedure in
this Virginia counterpart to the Maryland monitoring program was as
follows. A large volume sample (200 to 800 ml) was collected, filtered
onto a GF/F 2.5 cm filter on board the vessel with the addition of a
few drops of a magnesium carbonate suspension. The filter was held on
water ice until returning to the lab when it was frozen. In one case
(May 6, 1985), ice was not available and the samples were held in a dark
insulated, box until returning to the lab. At a later date the samples
were processed and data calculated as described above (Methods Heading 2)
for Spectrophotometric samples (i.e. by the method (Strickland and
Parsons, 1968) of grinding in 90% acetone, allowing to stand overnight in
the refrigerator, centrifuging and reading), with the exception that the
Lorenzen equation used a 664nm before acidification reading rather than
the 665.
For fluorometric readings, samples of either 5 or 10 ml were taken
in duplicate and processed as described above (Methods Heading 1) with
8 ml of the DMSO solvent on the vessel and read 3-7 days after the
cruise. Calculations were made without a correction for phaeo-pigments
although after acidification readings were taken for possible future use.
4. VIMS York River Plankton Monitoring
This monitoring program followed plankton-related parameters from
the Coast Guard Pier near the mouth of the York River for the
winter/spring bloom period and during the summer. Samples were collected
three times a week at high slack water. A surface sample was constructed
from equal parts of water from 1, 3, and 5 meters collected by bottle and
a bottom sample was collected by means of a pump. Water samples from
this study were placed in a cooler and returned to the laboratory within
30 minutes for processing. Chlorophyll samples were taken for this study
from the surface sample, July through September, 1985. Fluorometric
samples were taken in 5 ml duplicate samples on 25 mm GF/F filters,
extracted with DMSO and read 5-7 days later. Samples for Spectrophoto-
metric readings were in duplicate, 800 ml or less in volume, filtered
onto 47 mm GF/F filters with several drops of a saturated magnesium
carbonate suspension, and immediately ground with 90% acetone, held until
the next day in refrigeration, centrifuged and read. One or two
additional duplicate sets of samples were taken for Spectrophotometric
II-7
-------
analysis. One set was frozen for two weeks and one remained frozen for
4 to 8 weeks before analysis; the freezer temperature was -12 C.
Results
1. Comparison of solvents (DMSO and 90% acetone) for extraction by
fluorometry.
In the 1980 Chesapeake Bay data set, the DMSO extraction method
produced chl .a values under those test conditions which were equally as
good as those from the 90% acetone extraction with grinding. Using a
total of 136 pairs of observations, the two extraction methods produced
values which were statistically indistinguishable (Table II-3, lines 1
and 3), although there is less variation in the values uncorrected for
phaeophytin.
Table II-3. Comparison of two methods of extracting and calculating chl
a_ values. Values are (In DMSO - In 90% acetone).
Samples Mean Difference t PROB>|t| N
Between Extractions
1980 Chesapeake Bay Study
1. Corrected chl a. -0.05096 -1.05 0.2985 68
2. Phaeophytin 0.32321 4.88 0.0001 68
3. Uncorrected chl a. -0.002579 . -0.07 0.9450 68
4. Uncorrected vs 0.0853 2.09 0.041 68
corrected chl a_
•
1984-85 Maryland Chesapeake Bay Monitoring.
5. Uncorrected chl a 0.3208 11.4 0.00.01 95
Calculated phaeophytin values from the two solvents are highly
significantly different with the DMSO method producing higher values
(Table II-3, line 2). Uncorrected DMSO chl a_ values are significantly
higher than the corrected 90% acetone values (line 4). Thus DMSO seems to
extract chlorophyll J> (chl t>) more completely from these samples, i.e. an
increase in the chl _b interference would reduce the corrected chl ^
values and increase the calculated phaeophytin.
The comparison of the DMSO with the 90% acetone extraction methods
during the 1984-85 Maryland Chesapeake Bay Monitoring (Table II-3, line
5 and Figure II-l) proved to be highly significantly different with the
DMSO values being approximately 145% of the 90% acetone values. The
reason for this significant difference proved to be related to storage
conditions rather than analytical techniques. This can be best
illustrated by October 1984 samples where approximately half the samples
II-8
-------
DMSO VS ACETONE - FLUOROMETER
0
(fl
I
0
J
\
0
D
J
I
0
MARYLAND MONITORING
CHL UG/L ACETONE
Figure II-l. Maryland EPA Monitoring Program Samples: CBL-DMSO extract
measured by fluorometer compared to samples frozen and analyzed later at
VIMS by grinding in acetone for extraction and fluorometer determination.
Both data sets are calculated without phaeo-pigment corrections.
II-9
-------
DMSO VS GRINDING - FLUOROMETER
24
OCT 1964
22 -
20-
18 -
1« -
14-
12 -
10 -
8 -
6 -
4-
2 -
0
12
16
2O
24
CHL UGA GRINDING FUJOROMETER
Y-0 .923X- 0.311 Hi- Y-1.34X- 0.113
Figure II-2. October, 1984, Maryland samples frozen for two different
times. Grinding fluorometric analysis using Turner Model 111 ( O )
frozen 5 months, ( • ) using Turner Designs, frozen 11.5 months.
*•'•?.'':'-'
11-10
-------
were stored for 5 months whereas the other half were stored for 11.5
months (Figure II-2). The amount of measured chlorophyll clearly
declined with time.
2. Comparison of fluorometry with spectrophotometry.
2a. 90% Acetone with grinding.
Many of the Maryland Chesapeake Bay Monitoring samples were large
enough to produce 90% acetone extracts which could be read on the
spectrophotometer. Figure II-3 shows the relationship between the
fluorometric and spectrophotometric determinations on the same extracts
(90% acetone with grinding). Since the fluorometer was calibrated with
known chl a_ measured on the same spectrophotometer, one would expect to
see data like that of a calibration curve where the two values are
essentially identical. For these samples, which were stored for several
months and undoubtedly contained chlorophyll breakdown products, the •
fluorometric values averaged about 85% of the spectrophotometric value.
The two determinations are significantly different (Table II-4, line 1).
The fluorometric samples which are above about 15 jug 1 chl a_ on the
spectrophotometer seem to deviate more than those with < 15 jug. These
results may be dependent upon the breakdown products resulting from
storage but are unexplained at the time of this writing.
2b. DMSO/fluorometry compared to acetone/spectrophotometry
Data from the Virginia EPA Chesapeake Bay Monitoring are shown in
Fig II-4. The majority of these data show DMSO fluorometer values about
10% greater than those for the 90% acetone/spectrophotometric values and
are significantly different (Table II-4, line 2). The acetone/spec-
trophotometer samples were stored frozen for one to 3.5 weeks before
analysis whereas the DMSO/fluorometer samples were extracted on board the
research vessel and analyzed a few days later. Loss during storage to a
colorless breakdown product or a colored product wiT:h a lower absorbance
could produce the greater fluorometer values.
The VIMS York River Plankton Monitoring provided the opportunity to
carry out a similar comparison with all processing carried out by the
same laboratory personnel. Figure II-5a compares these data from the
DMSO fluorometer procedure with that of the 90% acetone grinding
spectrophotometer, all analyses carried out on fresh samples without a
storage period. The fluorometer values were significantly higher (Table
II-4, line 3) and appeared to be offset by a constant value rather than a
percentage of the spectrophotometric value. Subtracting a value of
1.643 from the fluorometric values (line in Fig. II-5a) produced data
which were not significantly different (Table II-4, line 4). Without
data between 0 and 5 iig 1 it is impossible to tell if in fact a zero
spectrophotometer reading could give a fluorometer reading of 1.6 jug 1 .
11-11
-------
(T
y
u
5
0
d
0
3
J
L
y
z
0
0
<
J
\
0
D
J
I
0
ACETONE - FLUOROMETER VS SPEC.
MARYUND MONITORING
45
40-
35-
30
25-
15-
10-
5-
to
i
20
30
40
CHL UG/L ACETONE SPECTROPHOTOMETER
Figure I1-3. Maryland EPA Monitoring Prograa Samples: samples frozen and
analyzed later at VIMS by grinding in acetone for extraction and analyzed
by fluorometer and spectrophotometer determination. Both data sets are
calculated without phaeopigment corrections. The spectrophotometric data
are calculated with the trichromatic equations of Jeffrey and Humphrey
(1975) for chl £, Jb, and c_.
11-12
-------
DMSO FLUOR vs ACETONE SPEC
y
t
2
0
£
0
D
L
0
W
IT
y
h
j
\
0
D
I
0
24
22
20-
18
16
14-
10-
8-
6-
4-
2-
VIRdNIA MONITORING
0
a a a
a o o QoD°
o o
a a
OG
0 2 ..,-4i 6 8 10 12 14 16 18
* ~ '*
CHL UG/L ACETONE SPECTROPHOTOMETER
— Y=X
Figure I1-4). Virginia EPA Chesapeake Bay Monitoring samples comparing
freshly extracted by DMSO fluorometrlc determinations (means of pairs),
with single 90Z acteone extracts with grinding after freezing. The 90Z
acetone extracts were read on the spectrophotometer and calculated by the
Jeffrey and Humphrey (1975) equations for chl &_t _b, and jc.
11-13
-------
o
33
YORK RIVER CHL. A, JULY-SEPT 85
OMSO-FIUOR. VS ACETONE-SPEC (FRESH)
3O -
25-
20 -
15-
1O-
a -
10
20
30
UC CHL A PER LITER - SPEC-FRESH
Y-X+1.643
v>
u
35
YORK RIVER CHL. A, JULY-SEPT 85
SPECTROPHOTOMETRIC FROZEN VS FRESH
30-
23 -
20-
13-
10-
10 20 3O
UC CHL A PER LITER - SPEC-FRESH
40
Figure II-5. The spectrophotometric data are calculated with the trichromatic
equations of Jeffrey and Humphrey (1975) for chl £, b^, and c. VIMS Coast
Guard Pier samples, July-Sept 1985.
A) Comparison of DMSO fluorometer, with 90% acetone with grinding
spectrophotometric data on fresh samples.
B) Effect of freezing; ( ) fresh samples, ( d ) frozen 2
weeks Y-0.789X+1.59, ( O ) frozen 4-6 weeks Y-0.699X+1.54
11-14
-------
Table II-4. Comparison of fluorometry with spectrophotometry for
determining chl a_ values. Values are (In Fluorometer - In
spectrophotometer).
Samples
Mean Difference
Between Methods
PROB>
N
1984-85 Maryland Chesapeake
Bay Monitoring.
1. Uncorrected chl a_ -1.116
1985 Virginia Chesapeake
Bay Monitoring.
2. Uncorrected chl a_ 0.4734
1985 Virginia York River
Plankton Monitoring.
3. Uncorrected chl <± 0.177
4. (Fluorometer -1.643) 0.000017
-5.11
15.8
0.0001
95
0.0001 177
4.42 0.0001 31
0.0004 0.99 31
3. Storage effects.
Early in the study we observed a difference between values
determined at CBL and those at VIMS. This persisted after complete
renovation and recalibration of equipment. During one trip between the
laboratories we made 12 replicates of DMSO plankton sample extracts, i.e.
the same water sample was divided and filtered onto 12 filters which were
placed in the DMSO tubes for extraction. Six of the tubes were
transferred to CBL and, the samples at VIMS and CBL were read the same
afternoon. The VIMS results were 3% higher numerically but not
significantly different from the CBL values (VIMS = 7.53, S.D. 0.52; CBL
= 7.30, S.D. 0.36; d.f. 10, t 0.819). As a result of this experience
we designed a simple frozen storage experiment (see methods). Results
are presented in Fig. II-5b. These data indicate a loss of chlorophyll
of about 20% during the first 2 weeks and an additional 10% loss in the
next 2-4 weeks. This loss could indicate either a partial conversion to
a colorless breakdown product or a combination with almost a complete
conversion to a colored form which should have an absorption coefficient
about 85% of that of chl £.
4. Presence of chlorophyll _b and c_
The spectrophotometric data allow chlorophylls _b and £ to be
calculated as well as a^ using the Jeffrey and Humphrey (1975) equations.
This was done for all the extracts with a chlorophyll concentration 0.2
jug/ml or above for the Virginia Chesapeake Bay monitoring program. Below
the concentration of 0.2 jug/ml extract values are unreliable (Lorenzen &
11-15
-------
Jeffrey, 1980). These values are plotted as ^:_b and a_:_£ ratios (Figure
II-6). Samples with low a:b ratios should have populations dominated by
Chlorophyceae (green algae), and samples with low a:c ratios should have
populations dominated by diatoms or dinoflagellates (see Table II-5).
There are no cell counts for these samples to verify these observations;
however, such analyses were attempted with the VIMS Coast Guard samples.
This attempt proved unsuccessful, presumably because the taxonomic
divisions of the counts were not detailed enough, i.e. categories were
too inclusive.
5. Precision of DMSO method.
The results from the 1980 Chesapeake Bay study indicate no signi-
ficant change in the determined values (P=0.99), nor in coefficient of
variation associated with the interval of storage (P=0.55). Presumably
if either additional materials were extracted with time or the extracted
pigment decomposed to colorless products during the storage period the
data would be more variable with longer storage/extraction time. Thus if
chl a_ is breaking down to phaeophytin £ or to other colored decomposition
products, this method registers the product as chl a_. It is therefore
practical to place the filters in the extraction tubes in the field and
read them in the lab at a later date.
Discussion
The July 1980 EPA Chesapeake Bay study showed to our satisfaction
that DMSO:acetone:water (9:9:2) was a satisfactory solvent when compared
to 90% acetone with grinding. The comparison was made with fluorometric
determinations uncorrected for phaeo-pigments. The main advantages of
this method were ease of sampling handling and storage (no grinding,
refrigeration, dilution). The samples are filtered, the filter placed in
solvent to extract, and the extract is decanted into the fluorometer tube
for the reading. The extracting sample can be stored at room temperature
for several weeks without affecting the results. This approach gives one
a value which amounts to chl a_ plus phaeo-pigments (including any which
were produced during storage), and may not be appropriate if phaeo-
pigment values are desired, however, it may be a perfectly adequate index
of phytoplankton biomass, i.e. living plus recently dead (or eaten)
phytoplankton.
It is apparent from a literature review that accessory pigments,
especially chlorophyll _b, interfere with both the fluorometric and the
spectrophotometric determination of phaeo-pigments and, conversely, the
presence of phaeo-pigments may interfere with the determinations of the
chlorophylls, especially chl a^ Chlorophyll b^has been shown to occur in
Virginia Bay Monitoring samples. Thus if either of these techniques is
used to measure pigments, compromises will have to be made. It is thus
apparent that if one really needs to know the amount of chlorophyll _a or
other pigments present, it (they) will have to be separated from
interfering substances prior to their determination. It is feasible to
do this with chromatographic procedures. Several investigators have
reported using thin layer chromatography (e.g. Garside and Riley, 1969;
Jeffrey, 1975). High Performance Liquid Chromatography (HPLC) is a
11-16
-------
VIMS BAY MONITORING
CHI A:B AND A:C APR1L-AUC 1985
5!
7
1
S
2
»
15-
14-
13-
12-
11 -
10-
0-
e-
7 -
8-
4-
3-
2-
1 -
C
a
a
a
a a
a a
°a °
_ a
° Q°° o ° ° a
0 am^f ° ° 0 ° °0
o m. o a ° oaa
o^i ° a
> 2 4 6 8 10 12 14 11
CHOOKOPMYU. A.-S
Figure II-6. The spectrophotometric data were calculated with the
trichromatic equations of Jeffrey and Humphrey (1975) for chl £, Jb, and
£ and the values below 0.2ug/ml extract were deleted. The remaining
values are plotted as a:b and a:c ratios.
11-17
-------
nj
s
-w <->
e
in -c:
s 8-
o
01
m
01
m
m
r—4
CJ
15 s
i—» O
01 *•
m "•
•i o>
>- — <
k. a
j= as
01 « k. B
i § j: a
-§.8 So,
o «• .a i
*j a. o ai
O. >- C 3
s
s-
&OI Ol
u nj
fe i-r
S 8-*
a -g -g
11-18
-------
better choice in that it can be automated to a large degree. Numerous
investigators have published using HPLC for chlorophyll determinations
(e.g. Abaychi and Riley, 1979; Brown, et al., 1981; Gieskes and Kraay,
1983; Goeyens, L. et al., 1982; Knight and Mantoura, 1985; Mantoura and
Llewellyn, 1983; Pearl et al., 1983; Shioi et al., 1983).
In summary, it appears that the fluorometric and spectrophotometric
methods for chlorophyll ^ estimations in general use have a fairly low
accuracy (optimistically perhaps within 30%) due to interference and
storage problems. A logical approach to chlorophyll £ estimation is to
use a fast simple extraction, such as the proposed DMSO approach which
involves a minimum of handling, possible storage at room temperature
and, thus, should improve precision no matter how the extract is
analyzed. The method of choice for extract analysis clearly is the use
of a chromatographic method to separate the pigments so that they can be
measured with less interference and greater accuracy. If this technique
isn't available, the individual investigator can use any or all of
several fluorometric and spectrophotometric methods to estimate the
chlorophyll pigments, including bulk breakdown products, at a sacrifice
in accuracy.
Comments on Interim Guidance on Quality Assurance/Quality Control (QA/QC)
for The Estuarine Field and Laboratory Methods.
The "Interim Guidance on Quality Assurance/Quality Control (QA/QC)
for The Estuarine Field and Laboratory Methods" (USEPA, 1985) provides a
standard operating procedure (SOP) for chlorophyll which essentially
paraphrases Strickland and Parsons (1972) for sample collection, and
processing and storage; it further recommends the fluorometric
method detailed in Strickland and Parsons (1972, Section IV.3.IV) based
on 90% acetone extractions, the implied use of the Turner Model 111
fluorometer and calibration by pigment extracts from a combination of
algal cultures.
Storage time: Strickland and Parsons (1972) suggest that filters with
chlorophyll samples may be stored "in the dark in a desiccator frozen to
-20 C but only for a few weeks. This procedure almost always leads to
low results and makes the extraction of chlorophyll more difficult;
filters should be extracted without delay if at all possible." Our
results agree with the loss of chlorophyll with weeks, e.g. 20% within 2
weeks. Our proposed solvent extraction technique using DMSO is easily
started immediately after filtering the sample in the field; we
recommend it over the acetone extraction because it eliminates the
problems of sample storage, grinding etc., while performing equally well.
Calibration: The Interim Guidance (USEPA, 1985) follows Strickland and
Parsons' (1972) recommendation that healthy cultures and a "mixture
of about equal amounts (by pigment) of Skeletonema costatum, Coccolithus
huxleyii, and Peridinium trochoidium be used as a source of
spectrophotometrically determined chlorophyll for calibration of the
fluorometer. It is our recommendation that commercially available
chlorophyll, not generally available in 1972, be used in the calibration.
Strickland and Parsons (1972) in fact state that calibration "must be
done on extracts from marine phytoplankton as pure chlorophyll a_ is
11-19
-------
difficult to obtain." Using pure chlorophyll should reduce
interlaboratory calibration differences and be an easily reproducible
frame of reference within a laboratory. Any potential advantage of
calibrating with a pigment mixture very similar to that of the sample
population quickly disappears in an estuarine environment having rapidly
changing pigment complements throughout the year. The use of chlorophyll
quality control (QC) samples available from the Environmental Monitoring
and Support Laboratory - Cincinnati (EMSL-Cincinnati) should be
incorporated into routine analyses programs.
The above comments generally apply also to the APHA (1985) Method
1001G2 which is essentially the same as Strickland and Parsons (1972).
The Interim Guidance should be more inclusive, or general, to include
other fluorometers such as the Turner Designs which is coming into
widespread use. For estuarine work, units of >ug per liter are more
appropriate than mg per cubic meter. The possibility of using HPLC to
separate the pigments before analysis should be both allowed and
encouraged. An evaluation of the costs of obtaining accurate and
informative data through automated HPLC techniques should be carried out.
Recommendations for the Chesapeake Bay Program
1. Take small samples 5-15 ml depending on chlorophyll concentration and
place them in the DMSO solvent on board the ship.
2. After 24 hours or upon return to port several days later, the samples
are read on the fluorometer and calculated without a phaeo-pigment
correction.
It should be recognized that this method although fast and easy,
will give the best data on euphotic zone samples which have few
chlorophyll decomposition products. Samples from near the bottom or
which contain sediments, fecal pellets, etc., will give values which are
inflated by the decomposition products.
Alternative Recommendation.
1. Take samples of 200-1000 ml and extract as in the above
recommendation.
2. Read the sample before and after acidification in a spectrophotometer
using a 1 cm cell only if the concentrations are above a fixed threshold
such as 0.25;ug/ml. For lower concentrations, small volume longer light
path (5 or 10 cm) cuvettes should be required.
3. An option to step 2 is to read the extract at multiple wavelengths as
well as before and after acidification and report all the pertinent data
so that users can make whatever calculations they wish, i.e. station
data, sample and extract volumes, and spectrophotometric readings and
length of light path.
11-20
-------
REFERENCES
Abaychi, J.K., and J. P. Riley. 1979. The determination of phytoplankton
pigments by high-performance liquid chromatography. Anal. Chim. Acta
107:1-11.
Adamski, J. M. 1976. Simplified Kjeldahl nitrogen determination for
seawater by a semiautomated persulfate digestion method. Anal. Chem.
43:1194-1197.
Afghan, B. K., Goulden, P. D., and Ryan, J. F. 1971. "Use of Ultraviolet
Irradiation in the Determination of Nutrients in Water with Special
Reference to Nitrogen." Tech. Bull. No. 40, Inland Waters Branch,
Department of Energy Mines and Resources, Ottawa, Canada.
APHA. 1985. Standard methods for the examination of water and wastewater.
American Public Health Association, Washington, DC 1268 pp.
Armstrong, F. A. J., P. M. Williams, and J. D. H. Strickland. 1966.
Photo-oxidation of organic matter in sea water by ultraviolet radiation,
analytical and other applications. Nature 211:481-483.
ASTM. 1979. Water. In: Annual Book of ASTM Standards, Part 31. Amer.
Soc. Test. Mat., Philadelphia.
Barnes, H. 1959. "Apparatus and Methods of Oceanography. Part One:
Chemical." George Allen and Unwin, Ltd., London.
Brown, L. M., B. T. Hargrave, and M. D. Mackinnon. 1981. Analysis of
chlorophyll a in sediments by high-pressure liquid chromatography. Can. J.
Fish. Aquat. Sci. 38:205-214.
Burnison, B. K. 1980. Modified dimethyl sulfoxide (DMSO) extraction for
chlorophyll analysis of phytoplankton. Can. J. Fish. Aquat. Sci.
37:729-733.
Conetta, A., A. Buccafuri, and J. Jansen. 1976. A semiautomated system
for the wet digestion of water samples for total Kjeldahl N and total P.
Am. Lab. 8:103-110.
D'Elia, C. F. 1983. Nitrogen determination in seawater. In: D. G.
Cappone and E. J. Carpenter [eds.], Nitrogen in the Marine Environment.
Academic Press, pp. 731-762.
D'Elia, C. F., P. A. Steadier, and N. Corwin. 1977. Determination of
total nitrogen in aqueous samples using persulfate digestion. Limnol.
Oceanogr. 22:760-764.
11-21
-------
Duncan, M. J., and P. J. Harrison. 1982. Comparison of solvents for
extracting chlorophylls from marine macrophytes. Bot. Mar. 25:445-447.
Ebina, J., T. Tsutsui, and T. Shirai. 1983. Simultaneous determination of
total nitrogen and total phosphorus in water using peroxodisulfate
oxidation. Water Res. 17:1721-1726.
Eppley, R. W., W. G. Harrison, S. W. Chisholm, and E. Stewart. 1977.
Particulate organic matter in surface waters off southern California and
its relationship to phytoplankton. J. Mar. Res. 35:671-696.
Fabbro, L. A., L. A. Filachek, R. L. lannacone, R. T. Moore, R. J. Joyce,
Y. Takahashi, and M. E. Riddle. 1971. Extension of the microcoulometric
determination of total bound nitrogen in hydrocarbons and water. Anal.
Chem. 43:1671-1678.
Faithfull, N. T. 1971. Automated simultaneous determination of nitrogen,
phosphorus, potassium and calcium on the same herbage digest solution.
20:41-44.
Froelich, P. N., and M. E. Q. Pilson. 1978. Systematic absorbance errors
with Technicon AutoAnalyzer II colorimeters. Water Res. 12:599-603.
Fuhs, G. W. 1971. Determinations of particulate phosphorus by alkaline
persulfate digestion.. Intern. J. Environ. Anal. Chem. 1:123-129.
Garside, C., and J. P. Riley. 1969. A thin-layer chromatographic method
for the determination of plant pigments in sea water and cultures. Anal.
Chim. Acta. 46:179-191.
Gibbs, C. F. 1979. Chlorophyll a and 'phaeo-pigments'. Aust. J. Mar.
Freshwater Res. 30:597-606.
Gieskes, W. W. C., and G. W. Kraay. 1983. Dominance of Cryptophyceae
during the phytoplankton spring bloom in the central North Sea detected by
HPLC analysis of pigments. Mar. Biol. 75:179-185.
Glibert, P. M., C. F. D'Elia, and Z. Mlodzinska. 1976. A semiautomated
persulfate oxidation technique for simultaneous total nitrogen and total
phosphorus determination in natural water samples. Woods Hole Oceanog.
Inst. Contrib. No. 3954.
Goeyens, L., E. Post, F. Dehairs, A. Vandenhoudt, and W. Baeyens. 1982.
The use of high pressure liquid chromatography with fluorimetric detection
for chlorophyll a determination in natural extracts of chloropigments and
their degradation products. Intern. J. Environ. Anal. Chem. 12:51-63.
Gordon, D. C. and Sutcliffe, W. H., Jr. 1974. Filtration of seawater
using silver filters for particulate nitrogen and carbon analysis. Limnol.
Oceanogr. 19:989-993.
Goulden, P. D., and D. H. J. Anthony. 1978. Kinetics of uncatalyzed
peroxydisulfate oxidation of organic material in fresh water. Anal. Chem.
50:953-958.
11-22
-------
Grasshoff, K., M. Ehrhardt, and K. Kremling. 1973. Methods of Seawater
Analysis, Second edition. Verlag-Chimie. Weinheim. 419 pp.
Henriksen, A. 1970. Determination of total nitrogen, phosphorus and iron
in fresh water by photo-oxidation with ultraviolet radiation. Analyst
95:601-608.
Hernandez, H. A. 1981. Total bound nitrogen determination by
pyrochemiluminescence. Am. Lab. 13:72-76.
Inskeep, W. P., and P. R. Bloom. 1985. Extinction coefficients of
chlorophyll a and b in N,N- dimethylformamide and 80% acetone. Plant
Physiol. 77:483-485.
Jeffrey, S. W. , and G. M. Hallegraeff. 1980. Studies of phytoplankton
species and photosynthetic pigments in a warm core eddy of the East
Australian Current. I. Summer populations. Mar. Ecol. Prog. Ser.
3:285-294.
Jeffrey, S. W., and G. F. Humphrey. 1975. New spectrophotometric
equations for determining chlorophyll a, b, cl and c2 in higher plants,
algae and natural phytoplankton. Biochem. Physiol. Pflanzen. 167:191-194.
Jeffrey, S. W., M. Sielicki, and F. T. Haxo. 1975. Chloroplast pigment
patterns in dinoflagellates. J. Phycol. 11:374-384.
Jirka, A. M., M. J. Carter, D. May, and F. D. Fuller. 1976. Ultramicro
semiautomated method for simultaneous determination of total phosphorus and
total Kjeldahl nitrogen in wastewaters. Env. Sci. Technol. 10:1038-1044.
Kalff, J., and E. Bentzen. 1984. A method for the analysis of total
nitrogen in natural waters. Can. J. Fish. Aquat. Sci. 41:815-819.
Kjeldahl, J. 1883. A new method for the determination of nitrogen in
organic matter. Z. Anal. Chem. 22:366-382.
Knight, R., and R. F. C. Mantoura. 1985. Chlorophyll and carotenoid
pigments in Foraminifera and their symbiotic algae: analysis by high
performance liquid chromatography. Mar. Ecol. Prog. Ser. 23:2.41-249.
Koroleff, F. 1970. Revised version of "Direct Determination of Ammonia in
Natural Waters as Indophenol Blue", Int. Counc. Explor. Sea, Paper C. M.
1969/C:9. ICES, Charlottenlund, Denmark.
Koroleff, F. 1976. Determination of ammonia. In: K. Grasshoff [ed.],
Methods of Seawater Analysis. Verlag Chemie, New York. pp. 126-133.
Langner, C. L., and P. F. Hendrix. 1982. Evaluation of a persulfate
digestion method for parlticulate nitrogen and phosphorus. Water Res.
16:1451-1454.
Lorenzen, C. J. 1967. Determination of chlorophyll and pheo-pigments:
spectrophotometric equations. Limnol. Oceanogr. 12:343-346.
11-23
-------
Lorenzen, C. J., and S. W. Jeffrey. 1980. Determination of chlorophyll in
seawater. UNESCO Technical Papers in Marine Science, No. 35. UNESCO, Paris
20 pp.
Lowry, J. H., and K. H. Mancy. 1978. A rapid automated system for the
analysis of dissolved total organic nitrogen in aqueous solutions. Water
Res. 12:471-475.
Mantoura, R. F. C., and C. A. Llewellyn. 1983. The rapid determination of
algal chlorophyll and carotenoid pigments and their breakdown products in
natural waters by reverse-phase high-performance liquid chromatography.
Anal. Chim. Acta 151:297-314.
Moran, R., and D. Porath. 1980. Chlorophyll determination in intact
tissues using N,N- Dimethylformamide. Plant Physiol. 65:478-479.
Moss, B. 1967. A note on the estimation of chlorophyll a in freshwater
communities. Limnol. Oceanogr. 12:340-342.
Nydahl, F. 1976. On the optimum conditions for the reduction of nitrate to
nitrite by cadmium. Talanta. 23:349-357-r
Paerl, H. W., J. Tucker, and P. T. Bland. 1983. Carotenoid enhancement
and its role in maintaining blue-green algal (Microcystis aeruginosa)
surface blooms. Limnol. Oceanogr. 28:847-857.
Parsons, T. R., Y. Maita, and C. M. Lalli. 1984. A manual of chemical and
biological methods for seawater analysis. Pergamon Press, New York 173
pp.
Reay, P. F. 1985. An improved determination of ammonia in Kjeldahl
digests and acidic solutions with a buffered berthelot reaction. Anal.
Chim. Acta 176:275-278.
Scheiner, D. 1976. Determination of ammonia and Kjeldahl nitrogen by
indophenol method. Water Res. 10:31-36.
SCORE (1966) Monographs on oceanographic methodology. Vol. 1.
Determination of photosynthetic pigments in seawater. UNESCO Press,
Paris. 69 pp.
Seely, G. R., M. J. Duncan, and W. E. Vidaver. 1972. Preparative and
analytical extraction of pigments from brown algae with dimethyl sulfoxide.
Mar. Biol. 12:184-188.
Shioi, Y., R. Fukae, and T. Sasa. 1983. Chlorophyll analysis by
high-performance liquid chromatography. Biochim. Biophys. Acta 722:72-79.
Shoaf, W. T., and B. W. Lium. 1976. Improved extraction of chlorophyll a
and b from algae using dimethyl sulfoxide. Limnol. Oceanogr. 21:926-928.
Smart, M. M., Reid, F. A. and Jones, J. R. 1981. A comparison of a
persulfate digestion and the Kjeldahl procedure for determination of total
nitrogen in freshwater samples. Water Res. 15:919-921.
11-24
-------
Snedecor, G. W., and W. G. Cochran. 1967. Statistical Methods. Sixth
ed., The Iowa State University Press.
Sol6rzano, L., and J. H. Sharp. 1980. Determination of dissolved
organic nitrogen in natural waters. Limnol. Oceanogr. 25:751-754.
Speziale, B. J., S. P. Schreiner, P. A. Giammatteo, and J. E. Schindler.
1984. Comparison of N, N-dimethylformamide, dimethyl sulfoxide, and
acetone for extraction of phytoplankton chlorophyll. Can. J. Fish. Aquat.
Sci. 41:1519-1522.
Stauffer, R. E., G. F. Lee, and D. E. Armstrong. 1979. Estimating
chlorophyll extraction biases. J. Fish. Res. Bd. Can. 36:152-157.
Stevens, R. J. 1976. Semi-automated ammonia probe determination of
Kjeldahl nitrogen in freshwaters. Water Res. 10: 171-175.
Strickland, J. D. H., and T. R. Parsons. 1968. A practical handbook of
seawater analysis. Fish.^Res. Bd. Can. Bull No. 167, 311 p.
Strickland, J. D. H., and T. R. Parsons. 1972. A Practical Handbook of
Seawater Analysis. Bull. Fish. Res. Board Can. 310 pp.
Suzuki, U., Y. Sugimura, and T. Itoh. 1985. A catalytic oxidation method
for the determination of total nitrogen dissolved in seawater. Mar. Chem.
16:83-97.
United States Environmental Protection Agency 1979. Methods for Chemical
Analysis of Water and Wastes. Off. Res. Devel. Cincinnati, Ohio.
EPA-60074-79-020.
United States Environmental Protection Agency 1985. Interim Guidance on
Quality Assurance/Quality Control (QA/QC) for The Estuarine Field and
Laboratory Methods. Office of Marine and Estuarine Protection (draft).
154pp.
Valderrama, J. C. 1981. The simultaneous analysis of total nitrogen and
total phosphorus in natural waters. Mar. Chem. 10:109-122.
Van Hall, C.E., Safranko, J., and Stenger, V.A. 1963. Rapid combustion
method for the determination of organic substances in aqueous solutions.
Anal. Chem. 35:315-319.
Webb, K. L. 1978. Nitrogen determination. In: "Coral Reefs: Research
Methods" (D. R. Stoddart and R. E. Johannes, eds.), pp. 413-419. Monographs
on Oceanographic Methodology, Vol. 5. SCOR/UNESCO, Paris.
Webb, K. L., W. D. DuPaul, W. J. Wiebe, W. Sottile, and R. E. Johannes.
1975. Enewetak (Eniwetok) Atoll: aspects of the nitrogen cycle on a coral
reef. Limnol. Oceanogr. 20:198-210.
Wood, A. M. 1979. Chlorophyll a:b in marine planktonic algae. J. Phycol.
15:330-332.
11-25
-------
Wood, L. W. 1985. Chloroform-methanol extraction of chlorophyll a. Can.
J. Fish. Aquat. Sci. 42:38-43.
Yentsch, C. S., and D. W. Menzel. 1963. A method for the determination of
phytoplankton chlorophyll and phaeophytin by fluorescence. "Deep-Sea Res.
10:221-231.
11-26
-------
Appendix I. Letter from B. Nowicky at the University of Rhode Island
summarizing her comparisons -of the TKN and TPN techniques as well as the
recovery, of caffeine-N using the TPN technique.
III-l
-------
University of Rhode bland, Narregansctt, Rhode k!axi 02882
Graduate School of Oceanography, Narragansett Bay Campus
February 6, 1986
Dr. Christopher D'Elia
Chesapeake Biological Laboratory
P. 0. Box 38
Solomons, Maryland 20688
Dear Dr. D'Elia:
I haven't forgotten your request for data comparing the Kjeldahl
technique with the Persulfate digestion for total nitrogen, I'm afraid
that locating that work (done some eight or nine years ago) is proving
more difficult than I expected. I've enclosed a brief table which may
be of some help. As the table shows, I first noticed that I got
consistently higher values for the Persulfate digestion than with the
Kjeldahl technique. When I checked my percent recovery of standard
additions of various organic compounds (urea, glycine, EDTA) to seawater,
I found I got better recovery with the Persulfate Technique. In. addition
I found that my precision was much better using a persulfate digestion.
The "caffeine recovery experiment" was done after Suzuki et al. (Mar.
Chem. !_£, (1985) 83-97) published an article questioning the ability of
the persulfate digestion to deal with ring nitrogen compounds. My decision
to switch to persulfate digestions was made after quite a lot of "playing
around" with the various techniques. Unfortunately, I never published
the data (or intended to) and it sits in my lab notebooks in disarray.
The table* I'm sending are some hits and pieces. I hope they're of use.
Sincerely,
Barbara Nowicki
BN/d
Enc.
III-2
-------
Six different samples were taken from the MERL experimental mesocosms
(salinity = 30 °/oo) and filtered (precompusted Glass fiber-filters). The
samples were then analysed using both Kjeldahl and Persulfate techniques.
Total dissolved nitrogen (ug-at L )
Tank #
5
5
5
7
7
7
Time Kjeldahl technique
9 a.m.
noon
3 p.m.
9 a.m.
noon
3 p.m.
10.9
10.8
11.7
12.0
14.4
11.3
Persulfate digestion
15.3
14.7
13.7
15.0
18.3
15.3
Kjeldahl technique - precision of duplicate estuarine samples.
Total dissolved
Sample Nitrogen (^g at IT1) x ± 1 s.d.
Brushneck Cove mouth
Brushneck Cove head
#1
#2
#1
#2
31.91
30.42
46.51
47.60
31.2 ' 1.05
47.1 0.8
Persulfate digestion - precision of six replicate estuarine samples from
the MERL mesocosms.
Total N Total P
x + s.d. x + s..d.
Unfiltered samples 60.3 ± 0.3 2.0 ± 0.08
Filtered samples 31.7 + 0.3 1.16 ± 0.04
III-3
-------
A check on percent recovery of various organic N compounds added to artificial
seawater using the persulfate digestion technique.
_
chart units % recovery
Compound (mean of 4 replicates) relative to NO^
10 uM NOj 11.63
10 uM Glycine 11.54 99%
10 uM Urea 11.55 99%
10 jiM Caffine 11.34 99%
III-4
-------
Appendix II. Raw data for TKN and TPN analysis performed on
continental shelf seawater spiked with standard.
Salinity Standard TKN TPN
% cone., Pk ht Cone., Recovery PK ht Cone., Recovery
0
0
,0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
BLANK
BLANK
BLANK
GLU
GLU
GLU
GLU
GLU
GLU
NH4
NH4
NH4
NH4
NH4
NH4
N03-
N03-
N03-
N03-
N03-
N03-
UREA
UREA
UREA
UREA
UREA
UREA
BLANK
BLANK
BLANK
GLU
GLU
GLU
GLU
GLU
GLU
NH4
NH4
NH4
NH4
NH4
NH4
N03-
N03-
N03-
N03-
N03-
N03-
UREA
UREA
UREA
UREA
0.0
0.0
0.0
25.2
25.2
50.4
50.4
75.5
75.5
15.0
15.0
45.0
45.0
75.0
75.0
25.0
25.0
50.0
50.0
75.0
75.0
26.8
26.8
53.6
53.6
80.4
80.4
0.0
0.0
0.0
25.2
25.2
50.4
50.4
75.5
75.5
15.0
15.0
30.0
30.0
75.0
75.0
25.0
25.0
50.0
50.0
75.0
75.0
26.8
26.8
53.6
53.6
16.0
15.8
14.1
28.0
30.6
39.6
38.9
59.3
53.7
24.3
21.8
39.3
38.8
48.1
50.3
14.4
14.5
13.9
15.9
13.6
13.3
30.0
29.6
41.5
40.8
48.7
53.1
11.3
12.1
12.5
29.5
27.3
32.5
43.5
59.1
56.9
23.6
27.8
44.0
45.8
54.5
58.6
16.1
24.4
15.5
15.3
22.7
18.3
31.6
27.6
45.6
50.1
2.22
1 .81
-1.74
27.24
32.66
51.43
49.97
92.51
80.83
19.53
14.32
50.81
49.76
69.15
73.74
-1.11
-0.90
-2.16
2.01
-2.78
-3.41
31.41
30.58
55.39
53.93
70.40
79.58
-7.58
-5.91
-5.07
30.37
25.78
36.63
59.56
92.09
87.50
18.07
26.83
60.60
64.36
82.50
91.05
2.43
19.74
1.18
0.76
16.19
7.02
34.75
26.41
63.94
73.32
1.08
1.30
1.02
0.99
1.23
1.07
1.30
0.95
1. 13
1.11
0.92
0.98
-0.04
-0.04
-0.04
0.04
-0.04
-0.05
1.17
1.14
1.03
1.01
0.88
0:99
1.21
1.02
0.73
1.18
1.22
1.16
1.20
1.79
2.02
2.15
1.10
1.21
0.10
0.79
0.02
0.02
0.22
0.09
1.30
0.99
1.19
1.37
7.3
8.2
9.4
25.4
27.8
44.4
43.2
63.4
66.2
17.4
24.7
40.5
41.8
59.0
59.4
27.3
28.5
50.3
50.2
64.5
64.7
26.6
26.8
53.4
50.4
70.7
67.3
10.1
10.8
8.2
26.6
26.2
46.7
48.6
66.2
64.9
20.7
20.1
31.6
27.3
64.1
64.1
27.6
20.9
48.3
48.3
68.3
68.5
26.8
25.9
48.4
48.8
0.00
0.00
0.60
21.50
24.60
46.20
44.70
71.00
74.60
20.60
11.10
41.20
42.80
65.20
65.80
24.00
25.50
53.90
53.80
72.40
72.60
23.10
23.30
57.90
54.00
80.50
76.00
1.50
2.40
0.00
23.00
22.40
49.10
51.60
74.50
72.80
15.30
14.50
29.50
23.90
71.80
71.80
24.30
26.10
51.20
51.20
77.20
77.50
23.20
22.10
51.30
51.90
0.85
0.98
0.92
0.89
0.94
0.99
1.37
0.74
0.92
0.95
0.87
0.88
0.96
1.02
1.08
1.08
0.97
0.97
0.86
0.87
1.08
1.01
1.00
0.95
0.91
0.89
0.97
1.02
0.99
0.96
1.02
0.97
0.98
0.80
0.96
0.96
0.97
1.04
1.02
1.02
1.03
1.03
0.87
0.82
0.96
0.97
III-5
-------
25
25
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
Ifl
75
75
75 '
UREA
UREA
BLANK
BLANK
BLANK
GLU
GLU
GLU
GLU
GLU
GLU
NH4
NH4
NH4
NH4
NH4
NH4
N03-
N03-
N03-
N03-
N03-
N03-
UREA
UREA
UREA
UREA
UREA
UREA
BLANK
BLANK
BLANK
GLU
GLU
GLU
GLU
GLU
GLU
NH4
NH4
NH4
NH4
NH4
NH4
N03-
N03-
N03-
N03-
N03-
N03-
UREA
UR!A
UREA
UREA
UREA
80.4
80.4
0.0
0.0
0.0
25.2
25.2
50.4
50.4
75.5
75.5
15.0
15.0
45.0
45.0
75.0
75.0
25.0
25.0
50.0
50.0
75.0
75.0
26.8
26.8
53.6
53.6
80.4
80.4
0.0
0.0
0.0
25.2
25.2
50.4
50.4
75.5
75.5
15.0
15.0
45.0
45.0
75.0
75.0
25.0
25.0
50.0
50.0
75.0
75.0
26.8
§§:i
53.6
80.4
80.4
NA
57.5
14.4
13. 1
11.9
29.9
23.8
42.4
42.0
55. 1
50.8
22.5
22.8
41.9
40.3
51.3
53.0
11.3
12.6
15.5
13.6
14.9
16.3
25.1
27.1
43.3
26.8
58.8
56.8
10.5
8.0
11.9
37.0
28.6
44.1
46.5
50.4
54.0
26.3
25.9
43.0
42.6
55.8
54.1
15.1
16.0
16.8
24.1
17.0
16.2
30.0
44 :i
49.0
57.5
58.9
NA
88.75
-1. 11
-3.82
-6.33
31.21
18.49
57.27
56.43
83.75
74.78
15.78
16.40
56.23
52.89
75.83
79.37
-7.68
-4.87
1.18
-2.78
-0.07
2.85
21.20
25.37
59.15
24.74
91.46
87.29
-9.24
-14.46
-6.33
46.01
28.49
60.81
65.82
73.95
81.46
23.70
22.87
58.52
57.69
85.21
81.66
0.35
2.22
3.89
19.11
4.31
2.64
31.41
15:1?
71.03
88.75
91.67
NA
1. 10
1.24
0.73
1. 14
1. 12
1.11
0.99
1.05
1.09
1.25
1.18
1.01
1.06
-0,31
-0.19
0.02
-0.06
.00
0.04
0.79
0.95
1.10
0.46
1.14
1.09
1.83
1.13
1.21
1.31
0.98
1.08
1.58
1.52
1.30
1.28
1.14
1.09
0.01
0.09
0.08
0.38
0.06
0.04
1.17
!:?§
1.33
1.10
1.14
70.9
74. 1
10.1
17.6
13.0
30.0
31.7
51.4
46.3
66.2
67.5
23.1
22.9
40.7
42.7
70.7
66.2
28.3
27.6
47.8
47.5
71.5
71.5
28.6
27.6
48.1
47.8
71.1
70.2
1.0
1.9
2.3
31.4
28.9
54.7
52.5
67.7
70.0
23.7
25.8
49.3
42.8
72.4
70.2
30.3
32.4
50.2
51.2
72.6
72.6
28.0
?§:i
52.8
67.1
66.6
80.60
84.80
1 .40
11. 10
5.20
27.30
29.50
55.10
48.50
74.40
76.10
18.30
18.10
41.20
43.80
80.50
74.60
25.10
24.20
50.50
50.10
81.30
81.30
25.50
24.20
50.80
50.50
80.80
79.60
.0.20
1.30
.1.80
28.90
25.70
59.20
56.40
76.20
79.10
18.90
21.60
52.20
43.70
82.30
79.40
27.50
30.20
53.40
54.70
82.50
82.50
24.90
§i:i8
56^80
75.40
74.70
1.00
1.05
1.08
1.17
1.09
0.96
0.99
1.01
1.22
1.21
0.92
0.97
1.07
0.99
1.00
0.97
1.01
1.00
1.08
1.08
0.95
0.90
0.95
0.94
1.00
0.99
1.15
1.02
1.17
1.12
1.01
1.05
1.26
1.44
1.16
0.97
1.10
1.06
1.10
1.21
1.07
1.09
1.10
1.10
0.93
1:88
1.06
0.94
0.93
III-6
-------
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100 .
100
100
100
100
100
100
100
100
100
100
100
BLANK
BLANK
BLANK
GLU
GLU
GLU
GLU
GLU
GLU
NH4
NH4
NH4-
NH4
NH4
NH4
N03-
N03-
N03-
N03-
N03-
N03-
UREA
UREA
UREA
UREA
UREA
UREA
0.0
0.0
0.0
25.2
25.2
50.4
50.4
75.5
75.5
15.0
15.0
45.0
45.0
75.0
75.0
25.0
25.0
50.0
50.0
75.0
75.0
26.8
26.8
53.6
53.6
80.4
80.4
10.5
10.0
11.8
24.8
26.3
43.6
39.4
50.6
53.3
19.5
20.8
42.5
38.4
55.8
51.8
14.3
14.3
NA
NA
15.5
16.2
27.3
26.5
49.5
45.8
58.2
60.4
-9
-10
-6
20
23
59
51
74
80
9
12
57
48
85
76
-1
-1
1
2
25
24
72
64
90
94
.24
.29
.53
.57
.70
.77
.01
.37
.00
.52
.23
.48
.93
.21
.87
.32
.32
NA
NA
.18
.64
.78
.12
.07
.36
.21
.80
0
0
1
1
0
1
0
0
1
1
1
1
-0
-0
0
0
0
0
1
1
1
1
.82
.94
. 19
.01
.98
.06
.63
.82
.28
.09
. 14
.02
.05
.05
NA
NA
.02
.04
.96
.90
.34
.20
.12
.18
13. 1
16.6
14.2
35.9
30.8
49.6
50.6
72.9
71.5
25.1
27.5
46.2
46.7
68.3
72.1
31.0
36.0
48.6
50.5
69.4
72.4
31.8
28.6
46.0
50.1
65.1
64.1
5.
9.
6.
34.
28.
52.
53.
82.
81.
20.
23.
48.
48.
76.
81.
28.
34.
51.
53.
78.
82.
30.
25.
47.
53.
72.
71.
00
60
40
70
00
50
80
80
00
60
70
10
70
80
80
30
80'
20
70
30
20
00
20
80
10
70
40
1.38
1. 11
1.04
1.07
1.10
1.07
1.37
1.58
1.07
1.08
1.02
1.09
1.13
1.39
1.02
1.07
1.04
1.10
1.12
0.94
0.89
0.99
0.90
0.89
III-7
-------
Appendix III.
Salinity Standard
%
0 glutamic acid
0 ammonia
0 nitrate
0 urea
25 glutamic acid
25 ammonia
25 nitrate
25 urea
50 glutamic acid
50 ammonia
50 nitrate
50 urea
75 glutamic acid
75 ammonia
75 nitrate
75 urea
100 glutamic acid
100 ammonia
100 nitrate
100 urea
Regression curves for TKN and TPN analyses
performed on continental shelf seawater spiked
with standard.
Method Intercept SEM Slope
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
TKN
TPN
SEM
15
7
15
9
15
8
16
7
11
9
16
9
13
8
12
-0
13
13
14
12
12
11
11
11
13
4
13
5
11
3
12
3
11
14
11
14
11
14
11
14
. 11
.97
.99
.42
.29
.87
.24
.99
.94
.17
.44
.25
.41
.17
.81
.71
.44
.40
.19
.54
.52
.77
.98
.96
.21
.04
.12
.10
.54
.64
.28
.83
.31
.27
.44
.66
.35
.47
.09
.32
1
0
0
1
0
1
0
1
1
0
3
0
1
1
1
1
1
1
1
1
0
1
2
1
2
1
1
2
1
1
1
1
1
1
1
0
0
1
1
0
.217
.775
.881
.187
.447
.032
.960
.150
.949
.711
.230
.758
.982
.524
.507
.593
.206
.281
.068
.538
.689
.702
.884
.575
.659
.951
.821
.139
.801
.423
.868
.762
.041
.098
.134
.828
.597
.137
.222
.909
0
0
0
0
-0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.529
.740
.462
.678
.020
.768
.446
.772
.587
.747
.593
.724
.092
.791
.604
.008
.537
.706
.531
.719
.036
.763
.527
.704
.570
.907
.593 '
.902
.106
.939
.604
.826
.558
.750
.586
.733
.066
.738
.623
.624
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.027
.017
.021
.028
.010
. 123
.020
.024
.044
.016
.083
.020
.045
.035
.037
.034
.027
.029
.026
.037
.016
.039
.061
.033
.060
.044
.044
.051
.041
.032
.040
.037
.023
.025
.027
.020
.014
.026
.026
.019
0.
0.
0.
0.
-0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
991
998
993
994
605
997
993
997
981
998
937
998
611
993
989
996
991
994
992
991
661
991
956
992
963
992
982
989
699
996
985
993
994
996
993
998
902
996
994
997
III-8
-------
Appendix IV. Tables from literature comparing precision of the total N
determinations by TKN and TPN.
(A) Seawater field samples. (D'Elia et al., 1977)
Concentration
*n
20
20-40
40-60
60-80
80-100
100-120
= # pairs
TPN
(uM) Mean (uM)
14.2
26.9
50.7
70.9
88.2
110.9
of samples analyzed
TKN + N03 and NO^
N*
23
14
11
20
12
16
CV%
8.7
5.9
8.6
5.2
3.2
3.7
Mean (uM)
14.3
27.1
47.3
70.1
____
N (pairs)
12
12
3
3
— — —
-N
CV%
5.3
6.9
7.3
2.2
— — —
(B) Standard samples (NHj-N) (Smart et al., 1981)
(3 samples analyzed -for each measurement)
TPN
Concentration
(iiM)
0.16
0.36
0.51
0.81
1.12
1.22
1.42
1.76
2.20
2.42
Mean
(mg-lT1)
0.17
0.39
0.49
0.83
1.08
1.21
1.51
1.84
2.17
2.48
CV%
20.05
4.07
2.22
7.85
4.24
3.33
3.04
4.69
2.02
4.85
Concentration
(uM)
0.10
0.20
0.30
0.60
0.80
1.20
1.40
1.60
2.00
2.40
TKN
Mean
(mg-lT1)
0.11
0.57
0.36
0.53
0.66
1.28
1.30
1.72
1.88
2.83
CV%
25.52
10.84
6.16
4.66
16.91
1.10
3.81
14.36 .
2.55
5.35
III-9
-------
(C) Freshwater field samples. (Smart et al., 1981)
(3 samples analyzed for each measurement)
TPN TKN
Sample Sites
Mean
(mg-lT1)
CV%
Mean
(mg-lT1)
cv%
Bear Creek above site 0.22 5.72 0.18 10.65
Silver Fork Creek 0.41 6.49 0.36 19.29
Mississippi River 0.80 6.22 0.55 5.79
Salt River 0.76 3.23 0.59 25.31
Hinkson Creek 0.69 4.46 0.61 9.90
Ted Shanks Marsh No. 8 1.05 2.28 0.61 25.25
Bear Creek Below Site 0.82 9.44 0.72 7.37
Ted Shanks Marsh No. 2 1.20 5.11 0.75 11.24
Cedar Lake 1.10 6.04 0.87 2.89
LeFevre Pond 4.83 6.88 4.39 9.49
111-10
-------
-------
APPENDIX E
RESULTS OF EPA AUDIT SWP481 PERFORMED BY CBL
A CHECK OF ACCURACY FOR DISSOLVED NITROGEN AND PHOSPHORUS
-------
-------
-------
25. March 1987
Dr. Robert Magnien
Office of Environmental Programs
Water Management Administration
Dept. of Health and Mental'Hygiene
201 W. Preston St.
Baltimore, Md. 21201
Dear Rob:
I am enclosing the results of quality control samples from EPA
unknowns WP481 performed by CBL in conjunction witlh the February 1987
mainstem samples. The actu-al concentrations of these unknowns were knc
only to myself and I had no part in the analyses.
Nutrient CBL EPA 9556 C.I. reporl
by EPA
Ammoni&-N 0.281 0.28 0.23-0.33
Nitrate-N 0,142 0.14 0.11-0.17
Orthophosphate-P 0.045 0.05 0.04-0.06
Total Kjeldahl-N 0.34 0.32 0.18-0.48
*Alkaline Persulfate-N 0.311 —
Total-P 0.107 0.10 0.07-0.13
*Alkaline Persulfate-P 0.104 --
All concentrations are reported in mg/1
Alkaline persulfate N and P were also performed on these unknowns
and the results are reported above. Again, in1eacli case, the values
obtained by the different method? are nearly identical.
These results will become part of our continuing QA/QC program for
1987. We are all very pleased with the results aad should you have any
questions, please call us at your convenience.
Sincerely yours,
Carl F. Zimmermann
cc: Dr. C.F. D'Elia
Mr. R. Batiuk
Ms. B. Fletcher
Nutrient Analytical Services file
-------
-------