CHEM 2080 Lecture Notes - Lecture 99: Conjugate Acid, Potassium Chloride, Galvanic Cell
140 views9 pages
:
-If any of the analyte wasn't transferred from a flask after separating it from other
components, there will be an error in the calculations.
-Calibration clock is a graph where you put peak height (y axis) vs analyte
concentration (in parts per million).
- - Certified Reference Material, shows if the results are accurate or not. Not
everything has a CRM, so you cant always check if the results are accurate.
-If there is no CRM, try using other methods and look if you get to the same (or
close) answer.
-Read 3.1 - 3.2 for more info about uncertainty. (review from high school).
- is how accurate the results are
- is how close the results are (they can be far from the real answer but
precise
-If we have to pick one, Accuracy is preferable.
-How to calculate central estimates: for the numbers 1,2,3,4,5,6,7,8,9.
-: (average) add all and divide by 9 = 5.
-: (the middle number) = 5, if even number of results add two mid
numbers and divide by 2.
-You can add variances (the square of standard deviation (SD)) but cannot add SD.
-: subtract the average from each number. Square each
subtraction and add all the squares. Divide by the number of numbers - 1. Square root.
- !: 100 x SD/mean = in %
-Types of errors: "(systematic)and #(random)
-" errors - You can see where you made the error and why. Can be
corrected
find more resources at oneclass.com
find more resources at oneclass.com
-Types of Determinate errors: #nstrumental Error, Method Error, Personal Error,
Constant Error and Proportional Error.
-# error for example if you use a dirty burette.
- error example is if you use a wrong indicator for a substance.
- error for example is when you don't line your eyes on the meter.
- error is when the difference between measured value and true value is
the same, its also called an absolute error. Use large amounts of analyte to minimize
error.
-$ error increases as the amount of analyte increases. Usually happens
when the analyte has contamination.
-We can fix Constant Errors by a %& experiment. A blank measurement is the
analysis of everything the same except its without the analyte. (Its like a control
experiment).
-# errors cannot be corrected. Effect - scattered values around the
mean.
-The statistics samples and population. Mean Standard
dev
-Finite number of measurements is called a $ x(with_on top) s
-Infinite number of measurements is called a $$ µ (true σ (true
-Properties of Gaussian distribution Mean) std v)
-Mean is in the central
-Symmetric (bell-shaped) gaussian equation
-50% below the mean, 50% above the mean
-4768 bulbs.
-Range sqrt(30) - 1100
-Random error
-2 variable
-µ and σ
-Slide 28 correction, against “N” is suppose to be against “z”.
find more resources at oneclass.com
find more resources at oneclass.com
-' = σ^2
-Similarity s^2 = variance
-When we combine two distributions, the variances are additive ( not standard
deviations)
-For different random error processes (1,2,3...n) that contribute to a measurement
error, the total variance.
-σ = sqrt(σ1^2 + σ2^2 + σ3^2 + ... σn^2)
-Coefficient of variation = σ/µ ≈ (s/x)*100%
-Pooled standard deviation is doing standard deviation with sets (see slide 35).
-Sample mean(x) can be measured, but the true mean (population, µ) cannot be
measured, but we can calculate it using the sample mean.
-Degree of freedom is n-1.
-Applications of confidence level: Comparison with a true value and Comparison
with two experimental mean.
-Example on slide 40 is a systematic error.
-T test is for accuracy. Tcalc > Tcritical = yes, there is a significant difference.
-Is there a significant difference (at 95% level) between precisions of the two labs?
Do F test.
-Is there a value that should be rejected from the results (off away from 95%
interval).
-Q test: Q value = gap/range for numbers (,1,2,2,3,3,) gap = 6-3 = 6-
1
-Pg. 67 examples. For errors (not on test).
-Everything in the sample except the analyte is a *. This refers to blank
method.
-$+ is the ability to distinguish the analyte from other compounds in the
sample.
-The range where a method can be reliably used is called a .
-Linear range is not the same as dynamic range.
find more resources at oneclass.com
find more resources at oneclass.com