
solve exponential equations using logarithms in base 2, 10 or e, graph exponential and logarithmic functions as inverses, and find the inverse of an exponential function.
Instruction
Logarithms exist to answer the question “how long?” for exponential models. The strategy: isolate the power, then take a logarithm.
Example 1. Solve 5 · 2t = 80. Divide by 5: 2t = 16, so t = log2 16 = 4. No calculator needed when the numbers cooperate.
Example 2. Solve 3 · 100.2t = 600. Divide by 3: 100.2t = 200. By definition, 0.2t = log 200, so t = 5 log 200 ≈ 11.505. Check: 100.2 × 11.505 = 102.301 ≈ 200.
The number e. If $1 is invested at 100% interest compounded n times a year, it grows to (1 + 1/n)n in a year. As n grows this approaches e ≈ 2.71828, the result of compounding continuously. Continuous growth at rate r is modeled by A = Pert, and its logarithm is the natural log, ln. At 4% continuous interest, $1,000 grows to 1,000e0.4 ≈ $1,491.82 in 10 years, slightly more than the $1,480.24 from 4% compounded yearly.
Example 3. When does 1,000e0.04t reach $3,000? Divide: e0.04t = 3, so 0.04t = ln 3 and t = ln 3/0.04 ≈ 27.47 years. This is the pattern the standard names: the solution to abct = d is t = logb(d/a)/c, evaluated with technology (A1.CED.A.4 asks you to derive it, and it takes two steps).
Doubling time. At 7% annual growth, solve 1.07t = 2: t = log 2/log 1.07 ≈ 10.24 years. The “rule of 70” (70 ÷ 7 = 10) is a quick approximation that comes from ln 2 ≈ 0.693.
Inverse functions (A2.BF.A.2). f(x) = 2x and g(x) = log2 x undo each other: g(f(x)) = x for every x, and f(g(x)) = x for every x > 0. Swapping inputs and outputs reflects a graph across y = x, so the graphs are mirror images (A2.BF.A.3). The exponential graph has y-intercept (0, 1) and horizontal asymptote y = 0; the log graph has x-intercept (1, 0) and vertical asymptote x = 0. The exponential rises faster and faster; the logarithm rises slower and slower but without bound.
Finding an inverse (A2.BF.A.2). For f(x) = 3 · 2x + 1, solve f(x) = c: 2x = (c − 1)/3, so x = log2((c − 1)/3). The inverse is f−1(x) = log2((x − 1)/3), defined for x > 1, which is the range of f. Check: f(x) = 10 gives x = log2 3 ≈ 1.585.
Graphical solutions (A2.REI.B.3). Some equations, like 2x = x + 3, cannot be solved with logarithms alone. Graph both sides and read the intersections: there are two, near x ≈ −2.86 and x ≈ 2.44.
📖 Read for this lesson
The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.
Formative check
Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.
What is the solution of 3 · 100.2t = 600?
The solution of 1,000e0.04t = 3,000 is t = ln / 0.04 ≈ years.
The graph of y = log2 x has a horizontal asymptote at y = 0.
A medication level is M(t) = 400(0.8)t mg, t in hours. Find the inverse function t(M), use it to find when the level reaches 50 mg, and confirm by graphing y = M(t) and y = 50. State the domain of your inverse in context.
How confident are you that you can solve exponential equations with logarithms and relate exponential and log graphs as inverses?
Practice
Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.
Printable version: this unit’s practice workbook (PDF)
Fluency
Build speed and accuracy with the core skill.
- Solve for x: 1.08x = 3.5. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 3.5 / log 1.08 ≈ 16.278Take the logarithm of both sides and use the power property. - Solve for x: 10x = 250. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 250 / log 10 ≈ 2.398Take the logarithm of both sides and use the power property. - Solve for x: 1.05x = 7. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 7 / log 1.05 ≈ 39.883Take the logarithm of both sides and use the power property. - Solve for x: 2x = 3.5. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 3.5 / log 2 ≈ 1.807Take the logarithm of both sides and use the power property. - Solve for x: 5x = 50. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 50 / log 5 ≈ 2.431Take the logarithm of both sides and use the power property. - Solve for x: 1.08x = 50. Give an exact answer using logarithms and a decimal to the nearest thousandth.
Check your answer
Answer: x = log 50 / log 1.08 ≈ 50.831Take the logarithm of both sides and use the power property. - Find the inverse of f(x) = 4x.
Check your answer
Answer: f−1(x) = (1/4)xSwap x and y, then solve for y. - Find the inverse of f(x) = −6x − 2.
Check your answer
Answer: f−1(x) = −(1/6)x − 1/3Swap x and y, then solve for y. - Solve 1,000e0.05t = 2,500.
Check your answer
Answer: t = ln 2.5/0.05 ≈ 18.33 - Find the inverse of f(x) = 2 · 3x − 4 and state its domain.
Check your answer
Answer: f−1(x) = log3((x + 4)/2), for x > −4
Application
Use the skill in context. Show your reasoning.
- At 5.5% annual growth, how long does an investment take to double?
Check your answer
Answer: ln 2/ln 1.055 ≈ 12.95 years - A medication level is 400(0.8)t mg after t hours. When does it reach 50 mg?
Check your answer
Answer: t = ln(1/8)/ln 0.8 ≈ 9.32 hours - With P(t) = 3,954,653(1.00634)t (Missouri, t years after 1950), in what year does the model reach 7 million?
Check your answer
Answer: t ≈ 90.4, around 2040
Challenge
Stretch problems. Expect to think before you write.
- Solve 2x = 3x − 1 exactly and to three decimals.
Check your answer
Answer: x = ln 3/(ln 3 − ln 2) ≈ 2.710x ln 2 = (x − 1) ln 3.
Review
Keep earlier skills sharp.
- Solve: 9 / (x − 3) = 4 / (x + 4)
Check your answer
Answer: x = −48/5Cross-multiply, solve, and check the solution does not make a denominator zero. - Solve: 3x2 − x − 6 = 0
Check your answer
Answer: x = (1 ± √73) / 6Use the quadratic formula and simplify the radical. A negative discriminant gives complex solutions.
Sixty-Six Years of Carbon Dioxide: A Record That Almost Fits an Exponential
NOAA GML, annual mean carbon dioxide at Mauna Loa Observatory, file co2_annmean_mlo.txt (created October 9, 2026), retrieved November 5, 2026 · Public domain (works of the United States government)
Why this reading. The workshop asks you to solve abct = d for real, and to derive and interpret an inverse. Solving for t is only interesting when someone actually wants the answer to “when?” This is the longest continuous record of atmospheric carbon dioxide in the world, from NOAA’s Mauna Loa Observatory. Values are annual means in parts per million, dry air mole fraction.
Table 1. Annual mean CO2, Mauna Loa (ppm)
| Year | 1959 | 1960 | 1970 | 1980 | 1990 | 2000 | 2005 |
|---|---|---|---|---|---|---|---|
| ppm | 315.98 | 316.91 | 325.68 | 338.76 | 354.45 | 369.71 | 379.98 |
| Year | 2010 | 2015 | 2020 | 2021 | 2022 | 2023 | 2024 |
| ppm | 390.10 | 401.01 | 414.21 | 416.41 | 418.53 | 421.08 | 424.61 |
| Year | 2025 | The full file carries every year from 1959 to 2025, with a stated uncertainty of 0.12 ppm. | |||||
| ppm | 427.35 | ||||||
Three models, three answers to the same question
Let t be years after 1960 and C(t) the annual mean. Take 450 ppm as the level to solve for — a round number, and one that appears in policy discussion.
Model A, the endpoints 1960 and 2025. The growth factor is (427.35/316.91)1/65 = 1.004610, so CA(t) = 316.91(1.004610)t, about 0.461% a year. Solve 316.91(1.004610)t = 450: t = ln(450/316.91) / ln 1.004610 = 76.2, which lands in 2036. Check the model in the middle: CA(30) = 363.8 against an actual 354.45 in 1990. It is 9.4 ppm high, and always high in the middle, because the early years grew more slowly than the endpoints imply.
Model B, a least-squares fit to all 66 years. Regressing ln C on t gives a factor of 1.004641, about 0.464% a year, with 1960 value 310.15. Solving for 450 from the 2025 value gives 11.2 years, so again about 2036. The two exponential models agree almost exactly on the date and still both miss: this model predicts 419.05 for 2025 against an actual 427.35.
Model C, the last twenty years. (427.35/379.98)1/20 = 1.005892, about 0.589% a year. Solving ln(450/427.35)/ln 1.005892 = 8.8 years puts 450 ppm in 2034.
A linear model, for comparison. Least squares on the raw ppm gives 1.686 ppm a year, reaching 450 in 2045. That is nine to eleven years later than any of the exponential models, and the reason is visible in the data: the gaps between decades are 8.8, 13.1, 15.7, 15.3, 20.4 and 24.1 ppm. The increments themselves are growing, so a constant-slope line will always fall behind. Exponential is the better shape here, even though none of these models is a physical theory of anything.
The inverse, and what it is for
Model A’s inverse answers “what year?” directly. From C = 316.91(1.004610)t, solve for t: t(C) = ln(C/316.91) / ln 1.004610, defined for C > 0, and meaningful for C above 316.91, the range of CA on t ≥ 0. The composition check is the whole point of an inverse: t(CA(65)) should return 65, and it does. Doubling time, from ln 2 / ln 1.004610, is 150.7 years — a number worth saying out loud, because it shows how slow 0.46% a year sounds and how fast 111 ppm in 66 years looks.
Where the model is wrong, and why that matters
Annual means hide a sawtooth. Mauna Loa readings rise and fall within every year by about 6 ppm as Northern Hemisphere plants take up carbon in summer and release it in autumn — the seasonal signal you will model with a sinusoid in Unit 3. An annual-mean model cannot tell you the value in May. More importantly, no exponential is a mechanism. This curve bends because of emissions, oceans and land uptake, all of which are policy- and physics-dependent, and any of which can change the curve inside a decade. A model that tells you 2036 is telling you what the last sixty-six years would give if nothing changed. Report the date and that sentence, or you have not reported the model.
Two more to solve. When does Model C reach 500 ppm? When did Model A pass 400 ppm, and how does that compare with the year the record actually crossed it? Both are one logarithm each, and both are worth checking against Table 1 rather than trusting.
Sources: NOAA Global Monitoring Laboratory, Trends in atmospheric carbon dioxide (gml.noaa.gov); NOAA GML, annual mean CO2 at Mauna Loa (data file) (gml.noaa.gov)