Exponential equations and functions

Learning objectives

Read the multiplier and its time interval together. A percent rate without a period is incomplete.

Constant factor versus constant difference

A linear relationship changes by a constant amount over equal input intervals. An exponential relationship changes by a constant factor. In f(t) = Abt, A > 0, b > 0,

A = f(0) is the initial value and b is the multiplier for one unit of t. If b > 1, the model grows; if 0 < b < 1, it decays. The case b = 1 is constant. If the coefficient or an added term changes sign, interpret the actual outputs rather than applying a growth slogan automatically.

An increase of p% gives multiplier 1 + p/100. A decrease of p% gives multiplier 1 − p/100. A multiplier of 1.07 means a 7% increase, while 0.93 means a 7% decrease. A multiplier of 1.7 means a 70% increase, not 7%.

Successive changes multiply. A 20% increase followed by a 20% decrease multiplies by (1.20)(0.80) = 0.96, not 1. Each percentage applies to a different starting amount.

Time intervals, doubling, and half‐life

When a factor b applies once every T time units, write

f(t) = Abt/T.

The exponent counts the number of growth or decay periods. The variables t and T must use the same units. Doubling and half-life models take the forms

f(t) = A 2t/T, f(t) = A (12)t/T.

For continuous time, a fractional number of periods is allowed by the model. For a count of completed cycles, the input may be restricted to integers.

To convert a multiplier from one interval to another, raise it to the ratio of interval lengths. If b applies over T units, the factor over s units is bs/T. Do not divide a compound percent rate by the number of subperiods and treat the result as exact.

An anchor is not always an initial value In f(t) = Ab(t−t0)/T, the coefficient A is the value at t = t0, not generally at t = 0. Setting the exponent equal to zero reveals the anchor. Increasing t by T then multiplies the output by b.

Build a model from a table or two values

For equal input gaps Δt, a pure exponential Abt has constant output ratio bΔt. If the gaps are unequal, the ratios must be interpreted over those unequal intervals. Given two positive values,

f(t2)f(t1) = bt2−t1.

First find the factor per unit, then substitute into either data point to find A. Equivalently, write an anchored model at a known data point and avoid the initial value entirely when it is not requested.

A finite table can fit many different functions. A pattern suggests an exponential model; when a problem states the function is exponential, that assumption makes the model choice justified. Do not claim that four displayed points prove a global growth rule without the model assumption.

A vertical offset changes what is multiplied. In f(t) = C + Abt, the exponential part is f(t) − C. Ratios of the total outputs need not be constant. When 0 < b < 1, the outputs approach C as t increases, but the exponential term does not become exactly zero at a finite time.

Solve equations and interpret thresholds

If both sides can be written as powers of the same positive base other than 1, equate the exponents. If the target is a power related to a given power, rewrite the target rather than finding the exponent itself. A ratio such as f(t + s)/f(t) often removes the initial coefficient completely.

For a numerical time that does not follow from recognizable powers, graph the model and the target horizontal line or use a numerical calculation. Keep the context in view: a continuous crossing time is not automatically the first completed year or cycle that meets a threshold. Test the neighboring integers in the original model.

Three common interpretation errors

The base is a factor, not a percent. Convert 1.12 to a 12% increase, and 0.88 to a 12% decrease.

The exponent defines the period. In A(1.12)t/4, 12% applies over four units of t.

An offset is not multiplied. In C + Abt, only the amount above or below C is scaled by b each unit.

Choose an informative equivalent form. If the question asks for a yearly factor, write the exponent as “years” and put the entire yearly factor in the base. If it asks for the value at a given year, anchor the exponent there. Algebraic rewriting is part of interpreting a model, not merely simplifying it.

15 worked examples

5.01. Interpret the coefficient and growth factor

A model gives N(t) = 650(1.04)t, where t is the number of months after observation begins. What do 650 and 1.04 mean?

Show worked solutionHide worked solution for example 5.01

Recognize the structure. Set t = 0 for the initial value and increase t by 1 to see the monthly multiplier.

Work it through. N(0) = 650(1.04)0 = 650. Also,

N(t + 1)N(t) = 1.04.

Thus the modeled amount starts at 650 and increases by 4% each month.

Answer: Initial amount 650; monthly growth factor 1.04, or 4% growth per month.

Check. After one month the model gives 676, which is 26 more than 650, and 26/650 = 0.04.

Avoid the trap. 1.04 is not a 104% increase. It represents the original 100% plus a 4% increase.

5.02. Create a depreciation model

An item’s initial value is $18,000, and a model assumes it loses 12% of its current value each year. Write its value V(t) after t years.

Show worked solutionHide worked solution for example 5.02

Recognize the structure. Losing 12% means retaining 88%, so the yearly multiplier is 0.88.

Work it through. Multiply the initial value by that factor once per year:

V(t) = 18000(0.88)t.

The model can allow real t ≥ 0 if time is continuous, or integer t if only year-end values are considered.

Answer: V(t) = 18000(0.88)t.

Check. V(1) = 15840, which is 2160 = 0.12(18000) below the initial value.

Avoid the trap. 18000(0.12)t would mean retaining only 12% each year, a much larger decline. A fixed subtraction each year would instead be a linear model.

5.03. Count half‐life periods

The amount of a substance is modeled by M(t) = 96(1/2)t/3 grams, where t is hours. How many grams remain after 9 hours?

Show worked solutionHide worked solution for example 5.03

Recognize the structure. Nine hours contain three periods of length three hours.

Work it through. M(9) = 96 (12)9/3 = 96 (12)3 = 968 = 12.

Answer: 12 grams.

Check. Successive three-hour amounts are 96, 48, 24, and 12 grams.

Avoid the trap. The exponent is t/3, not 3t. The factor 1/2 applies every three hours, not every hour.

5.04. Solve using an equal base

Solve 2x+1 = 32.

Show worked solutionHide worked solution for example 5.04

Recognize the structure. 32 = 25, so both sides can use base 2.

Work it through. 2x+1 = 25 x + 1 = 5 x = 4.

Answer: 4.

Check. 24+1 = 25 = 32.

Avoid the trap. The exponent is the entire expression x + 1. Matching x directly to 5 ignores its constant term.

5.05. Read an exponential rule from a table

The function f has the form f(t) = Abt. Use the table to determine f(t).

t 0 1 2 3

f(t) 5 15 45 135

Show worked solutionHide worked solution for example 5.05

Recognize the structure. The inputs are equally spaced, so consecutive output ratios reveal the per-unit factor.

Work it through. Each ratio is 3: 15/5 = 45/15 = 135/45 = 3. Therefore b = 3. At t = 0, f(0) = A = 5.

Answer: f(t) = 5 ⋅ 3t.

Check. The rule gives 5 ⋅ 33 = 135 at t = 3 and reproduces every listed value.

Avoid the trap. The first differences are 10, 30, and 90, not a constant. A linear rule is not consistent with this table.

5.06. Handle data points separated by several periods

An exponential function f(t) = Abt, with A > 0 and b > 0, satisfies f(2) = 80 and f(5) = 640. What is f(0)?

Show worked solutionHide worked solution for example 5.06

Recognize the structure. The output ratio covers a three-unit input increase, not a one-unit increase.

Work it through. Divide the known values:

b5−2 = 64080 = 8 b3 = 8 b = 2.

Then 80 = A(22), so A = 20.

Answer: f(0) = 20.

Check. 20 ⋅ 22 = 80 and 20 ⋅ 25 = 640.

Avoid the trap. Using 8 as the per-unit multiplier would compound the full three-unit factor every unit and fail the data.

5.07. Convert a multiyear factor to a yearly rate

A quantity is modeled by Q(t) = 800(1.08)t/3, where t is in years. What is the equivalent annual percent increase?

Show worked solutionHide worked solution for example 5.07

Recognize the structure. The 8% increase occurs over three years. The one-year factor is the cube root of 1.08.

Work it through. Rewrite the function as

Q(t) = 800 ((1.08)1/3)t.

The annual multiplier is (1.08)1/3. Subtract 1 and multiply by 100 to convert the factor to a percent.

Answer: 100((1.08)1/3 − 1)%, approximately 2.60% per year.

Check. Cubing the annual factor returns 1.08, the required three-year multiplier.

Avoid the trap. 8%/3 is only a rough simple-rate approximation. It does not reproduce the compound factor exactly.

5.08. Find elapsed time from a recognizable decay ratio

The mass of a sample is M(t) = 640(1/2)t/6 grams, where t is hours. When does the model give a mass of 80 grams?

Show worked solutionHide worked solution for example 5.08

Recognize the structure. 80 is 1/8 of 640, which is three factors of 1/2.

Work it through. Set the target and divide by 640:

(12)t/6 = 80640 = 18 = (12)3.

Thus t/6 = 3, so t = 18.

Answer: 18 hours.

Check. Three half-lives give 640→320→160→80, taking 3 ⋅ 6 = 18 hours.

Avoid the trap. The number 3 is a count of half-life periods, not the answer in hours. Convert the count back to the stated time unit.

5.09. Turn a monthly factor into an annual factor

The model P(t) = 1500(1.02)12t uses t in years. What is its annual growth factor, and approximately what is the annual percent increase?

Show worked solutionHide worked solution for example 5.09

Recognize the structure. One year contains twelve applications of the factor 1.02.

Work it through. Use the power rule:

P(t) = 1500 ((1.02)12)t.

The annual factor is (1.02)12 ≈ 1.26824, so the annual percent increase is approximately 100(1.26824 − 1) = 26.8%.

Answer: Annual factor (1.02)12; approximately 26.8% annual growth.

Check. Setting t = 1 multiplies 1500 by exactly that annual factor.

Avoid the trap. Twelve times 2% gives 24%, not the compounded annual change. Each month’s increase also grows during later months.

5.10. Interpret a shifted exponential anchor

For t in days, N(t) = 200 ⋅ 3(t−4)/5. What time does the coefficient 200 refer to, and how long does each tripling take?

Show worked solutionHide worked solution for example 5.10

Recognize the structure. Set the exponent equal to zero for the anchor; increase the exponent by 1 for one tripling.

Work it through. At t = 4, the exponent is 0, so N(4) = 200. Increasing t by 5 increases (t − 4)/5 by 1, multiplying the output by 3.

Answer: The modeled amount is 200 at day 4, and it triples every 5 days.

Check. N(9) = 200 ⋅ 3 = 600, while N(4) = 200.

Avoid the trap. The coefficient is not the value at day 0 here. The initial value is 200 ⋅ 3−4/5 because the exponent has a shift.

5.11. Recover an initial value from two noninitial values

Let f(x) = abx, where a > 0 and b > 0. If f(2) = 18 and f(4) = 162, what is a?

Show worked solutionHide worked solution for example 5.11

Recognize the structure. A ratio removes a and determines b2.

Work it through. ab4ab2 = 16218 = 9 b2 = 9.

Since b > 0, b = 3. Now ab2 = 18 gives 9a = 18, so a = 2.

Answer: 2.

Another route. The midpoint value is f(2) = ab2, and f(4) = a(b2)2. Thus a = f(2)2/f(4) = 182/162 = 2.

Avoid the trap. The condition b > 0 is part of the real exponential model. A negative base is not interchangeable with a positive base for arbitrary real inputs.

5.12. Use an interval ratio without knowing the initial amount

An exponential function g(t) = Abt, with A > 0 and b > 0, satisfies g(t + 4) = 5g(t) for every real t. What

is g(t + 10)g(t)?

Show worked solutionHide worked solution for example 5.12

Recognize the structure. The given four-unit ratio is b4 = 5. Express the ten-unit ratio as a power of it.

Work it through. g(t + 10)g(t) = b10 = (b4)10/4 = 55/2 = 255.

Both the initial coefficient and the specific starting time cancel.

Answer: 255.

Check. Ten units equal 2.5 four-unit periods, so the ratio is 52.5. This is greater than 25 and less than 125, as expected.

Avoid the trap. A ten-unit multiplier is not 5(10/4). Repeated proportional change is represented by a power, not linear scaling of the multiplier.

5.13. Change both the time unit and the exponential base

A model is N(t) = 50⋅4t/6, where t is minutes. The same model is written N = 50⋅2s/k, where s is seconds. What is k?

Show worked solutionHide worked solution for example 5.13

Recognize the structure. Convert minutes to seconds before matching powers of 2.

Work it through. Because t = s/60,

4t/6 = 4s/360 = (22)s/360 = 2s/180.

Matching the requested exponent gives k = 180.

Answer: 180.

Check. The original model quadruples every 6 minutes. That is two doublings, so each doubling takes 3 minutes, or 180 seconds.

Avoid the trap. Changing only the time unit would leave base 4. Changing only the base would leave minutes. Both conversions must be reflected in the exponent.

5.14. Separate a fixed background from an exponential part

A function has the form f(t) = c + abt, with a > 0 and 0 < b < 1. The table gives four values. What is c?

t

0

1

2

3

f(t)

14

10

8

7

Show worked solutionHide worked solution for example 5.14

Recognize the structure. The fixed offset cancels in differences, even though it prevents constant ratios of the total outputs.

Work it through. The first differences are −4, −2, and −1. In this model consecutive differences have ratio b, so b = (−2)/(−4) = 1/2. From f(0) = 14 and f(1) = 10,

c + a = 14, c + a2 = 10.

Subtracting gives a/2 = 4, so a = 8 and c = 6.

Answer: 6.

Check. 6 + 8(1/2)2 = 8 and 6 + 8(1/2)3 = 7, matching the remaining data.

Avoid the trap. The total outputs do not halve. Their excess above 6 halves: 8, 4, 2, 1.

5.15. Build a growth model, then enforce a whole‐number threshold

An exponential function f(t) = Abt, with A > 0 and b > 0, satisfies f(1) = 12 and f(3) = 48. What is the least nonnegative integer n for which f(n) > 200?

Show worked solutionHide worked solution for example 5.15

Recognize the structure. First identify the model. Then check neighboring integer inputs rather than rounding a crossing time.

Work it through. The ratio gives b2 = 48/12 = 4, so b = 2. From Ab = 12, A = 6. Therefore f(n) = 6 ⋅ 2n.

Evaluate consecutive powers near the threshold:

f(5) = 6 ⋅ 32 = 192 < 200, f(6) = 6 ⋅ 64 = 384 > 200.

Since the function is strictly increasing, all smaller nonnegative integers give values at most 192.

Answer: 6.

Check. The derived rule gives f(1) = 12 and f(3) = 48, so it satisfies both starting conditions.

Avoid the trap. The answer must be an integer that meets the strict inequality. Rounding a continuous crossing downward would choose a value that has not reached the threshold.

What this section should change in your approach

Read a factor and its interval as one piece of information. Use ratios to remove an unknown initial amount, and rewrite the exponent when units or anchors change. For an offset model, track the amount above the offset. For a whole-number threshold, verify the last failure and first success.

Section exit check

Attempt these five questions without the lesson or worked solutions. Give exact answers unless a decimal is requested. Open each model answer after attempting the question. These are learning checks, not an official score scale.

Topic practice

Mastery check

Explain why the base in Question 1 is not 0.15, why the two-year ratio in Question 3 is not a one-year ratio, and why Question 4 requires a root rather than division of a percent.